Non-point source pollution and waterlogging collaborative four-pre management and control system based on short temporary rainfall forecast

Abstract

The invention discloses a non-point source pollution and waterlogging collaborative four-pre management and control system based on short temporary rainfall forecast, which relates to the technical field of urban water environment management and comprises a short temporary rainfall forecast module, a non-point source pollution control module, a waterlogging management module, a four-pre management and control platform, an intelligent decision support system, a real-time monitoring and feedback system and a water ecological management module. By integrating meteorological monitoring, hydrological monitoring and environment monitoring data, high-precision short and temporary rainfall forecast is realized by using a machine learning algorithm, the operation strategies of the intelligent catch basin and the initial rain storage pond are dynamically adjusted to control non-point source pollution, and meanwhile, the waterlogging risk is predicted and preventive measures are taken in combination with real-time hydrological data. According to the technology, integrated management of forecasting, early warning, rehearsal and plans is achieved through a four-pre-management and control platform, an intelligent decision support system, a real-time monitoring and feedback system and a water ecological management module under the sponge city design concept are used in an auxiliary mode, and the efficiency and effect of urban water environment management are comprehensively improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of urban water environment management, and particularly relates to a collaborative four-prevention control system for non-point source pollution and waterlogging based on short-term and impending precipitation forecasting. Background Art

[0002] With the acceleration of the urbanization process, urban water environment management is facing unprecedented challenges. Urbanization has led to an increase in impervious areas, changing the characteristics of surface runoff and making cities more prone to waterlogging when encountering short-term heavy precipitation. At the same time, the problem of urban non-point source pollution has become increasingly prominent. Especially during the rainy season, due to rainwashing, pollutants on the urban surface (such as road sediments, agricultural non-point source pollution, etc.) will enter the water body with runoff, causing water quality deterioration. These problems not only affect the quality of life of urban residents but also pose threats to the ecological environment and public safety.

[0003] 1. Current situation of short-term and impending precipitation forecasting technology: Traditional short-term and impending precipitation forecasting mainly relies on observational data from meteorological stations and numerical weather prediction models. However, these methods often have problems such as insufficient forecasting accuracy and low update frequency, making it difficult to meet the requirements of urban fine management.

[0004] 2. Current situation of urban waterlogging management technology: Urban waterlogging management usually relies on the construction and maintenance of drainage systems and some emergency measures. However, these measures often respond passively, lacking predictability of precipitation events and active management of waterlogging risks.

[0005] 3. Current situation of non-point source pollution control technology: Non-point source pollution control technologies include source control, process control, and end control, etc. However, these technologies often operate independently, lacking systematic integration and collaborative management.

[0006] 4. Current situation of four-prevention control technology: Although the concept of "four-preventions" has been proposed and applied in multiple fields, in urban water environment management, especially for four-prevention control technology for short-term and impending precipitation events, there is still a lack of effective integration and implementation.

[0007] 5. Current situation of intelligent decision support systems: The application of existing intelligent decision support systems in urban water environment management is not extensive enough, and they mostly rely on static data and models, lacking the integration of real-time data and dynamic decision-making capabilities.

[0008] 6. Current situation of real-time monitoring and feedback systems: Although some real-time monitoring systems have been deployed to monitor the urban water environment, these systems are often limited to the monitoring of single parameters and lack the comprehensive monitoring capabilities of multiple parameters and multiple scales.

[0009] 7. Current status of water ecological governance technologies: Water ecological governance technologies have been applied to a certain extent in urban water environment management, but these technologies often focus on a single water body or region and lack holistic and systematic planning. Summary of the invention

[0010] The technical problem to be solved by the present invention is to provide a coordinated four-prediction control system for non-point source pollution and waterlogging based on short-term precipitation forecasts in response to the shortcomings of the background technology; by integrating meteorological monitoring, hydrological monitoring and environmental monitoring data, a machine learning algorithm is used to achieve high-precision short-term precipitation forecasts, and dynamically adjust the operation strategies of intelligent interception wells and initial rain storage pools to control non-point source pollution, while combining real-time hydrological data to predict waterlogging risks and take preventive measures; through the four-prediction control platform, integrated management of forecasts, early warnings, rehearsals and plans is achieved, assisted by an intelligent decision support system and a real-time monitoring and feedback system, as well as a water ecological governance module under the sponge city design concept, to comprehensively improve the efficiency and effectiveness of urban water environment management.

[0011] The present invention adopts the following technical solutions to solve the above technical problems:

[0012] A coordinated four-precautionary control system for non-point source pollution and waterlogging based on short-term precipitation forecast, comprising a short-term precipitation forecast module, a non-point source pollution control module, a waterlogging management module, a four-precautionary control platform, an intelligent decision support system, a real-time monitoring and feedback system, and a water ecological management module;

[0013] Among them, the short-term precipitation forecast module is used to use meteorological data and machine learning algorithms to achieve high-precision forecasts of short-term precipitation;

[0014] The non-point source pollution control module is used to adjust the operation strategy of the intelligent interception well and the initial rain storage pool according to the short-term precipitation forecast to control non-point source pollution;

[0015] The waterlogging management module is used to combine short-term precipitation forecasts and real-time hydrological data to predict waterlogging risks and take corresponding preventive measures;

[0016] Four pre-control platforms are used to integrate forecasting, warning, rehearsal and emergency plan functions to achieve coordinated control of short-term precipitation, non-point source pollution and waterlogging;

[0017] Intelligent decision support system, used to provide decision support based on data analysis, including emergency response, resource allocation and pollution control strategies;

[0018] Real-time monitoring and feedback system, which is used to deploy sensor networks to monitor water quality, water level and flow in real time, ensuring real-time data update and feedback;

[0019] The water ecological management module is used to integrate the sponge city design concept and improve the water environment quality through ecological restoration and water purification technology.

[0020] As a further preferred solution of the surface source pollution and waterlogging collaborative four - prediction management and control system based on short - term and imminent precipitation forecasting of the present invention, the short - term and imminent precipitation forecasting module realizes high - precision forecasting of future precipitation based on a deep learning model and dual - polarization radar data; uses deep learning models, especially convolutional neural network CNN and generative adversarial network GAN, to process and analyze meteorological data, so as to predict short - term and imminent precipitation events; specifically includes the following steps;

[0021] Step A1, data pre - processing: Read the horizontal reflectivity factor Z in the dataset h data, design different threshold segmentation algorithms, and eliminate precipitation events with less or no precipitation in the dataset

[0022] to obtain data meeting the requirements of heavy precipitation;

[0023] Step A2, establish a deep learning model: Use deep learning models, including ConvLSTM, RNN, Transformer, to directly predict the future rainfall rate using radar; the model needs to be able to handle complex non - linear spatio - temporal conversions to reduce the ambiguity of prediction; the specific calculation is as follows:

[0024] Horizontal reflectivity factor where D is the diameter of the particle, N(D) is the number concentration of particles per unit volume of air and per unit size, λ is the radar wavelength, D m is the maximum and minimum values of the particle diameter, K w is the dielectric constant of the particle, f h (π,D) is the backscattering amplitude in the horizontal polarization direction;

[0025] Differential reflectivity where Z v is the vertical reflectivity factor;

[0026] Specific differential phase shift where R e is the real part value of the integral part, f h (0,D) and f v (0,D) are the forward scattering amplitudes in the horizontal polarization direction and vertical polarization direction respectively;

[0027] Step A3, model accuracy evaluation: Quantitatively evaluate the prediction accuracy of the model, including comparing with the optical flow method, traditional deep learning methods and U - Net method, and evaluate the forecasting effect of the model on different time scales.

[0028] As a further optimized solution of the surface source pollution and waterlogging collaborative four-prevention control system based on short-term and imminent precipitation forecasting in the present invention, the surface source pollution control module controls surface source pollution through the collaborative operation of intelligent intercepting wells and first-flush storage ponds; using short-term and imminent precipitation forecasting data, dynamically adjusts the operation strategies of these two facilities to reduce the entry of pollutants into water bodies; specifically includes the following steps;

[0029] Step B1, calculation of the storage pond volume: According to short-term and imminent rainfall forecasting, determine the rainfall duration and rainfall intensity; use the terrain and land use conditions to determine the catchment area and runoff coefficient; calculate the required volume of the storage pond to ensure effective control of surface source pollution during rainfall;

[0030] Calculation of the storage pond volume: Considering factors such as short-term and imminent rainfall, ground scouring, sediment, etc., the calculation formula for the storage pond volume is expressed as: Where: V is the effective volume of the storage pond; F is the catchment area; ψ is the runoff coefficient, considering the influence of rainfall and ground conditions; Δt is the rainfall duration, determined according to short-term and imminent rainfall forecasting; r is the rainfall intensity, also determined according to short-term and imminent rainfall forecasting; ψz is the comprehensive rainfall runoff coefficient of the construction site;

[0031] Step B2, calculation of pollutant concentration: Determine the initial pollutant concentration at the start of rainfall through monitoring; calculate the change in pollutant concentration at different time points according to the rainfall intensity and rainfall duration; adjust the operation strategy of the sewage treatment plant to adapt to the change in pollutant concentration under different rainfall conditions;

[0032] Calculation of pollutant concentration: Considering factors such as the amount of initial rain allowed to be discharged to the sewage treatment plant, the monitored concentration of pollutants in rainwater, the allowed concentration for discharging into the river, the rainfall duration, etc., the calculation formula for pollutant concentration is expressed as:

[0033] Where: C t is the pollutant concentration at time ; C0 is the initial pollutant concentration at the start of rainfall; k is the scouring coefficient, related to the ground scouring and sediment removal rate; r is the rainfall intensity; t is the rainfall duration; C bg is the background concentration, that is, the pollutant concentration during non-rainfall period;

[0034] Step B3, adjustment of the operation strategy of the intelligent intercepting well: According to short-term and imminent precipitation forecasting, the control system of the intelligent intercepting well will adjust the opening and closing degrees of the sewage interception gate and the rainwater gate to control the inflow of pollutants. The specific strategy includes closing the sewage interception gate before predicting a large amount of precipitation to intercept pollutants, and at the same time opening the rainwater gate to reduce the risk of waterlogging;

[0035] Step B4, initial rain storage tank operation strategy adjustment: The initial rain storage tank will adjust its operation mode according to the precipitation forecast. Before the precipitation event is predicted, the storage tank can be emptied in advance to leave storage space for the upcoming rainwater. During the precipitation period, the storage tank will collect the initial rainwater and process it after the rain to reduce the emission of pollutants.

[0036] As a further preferred solution of the non-point source pollution and waterlogging coordinated four-prevention control system based on short-term precipitation forecast of the present invention, the waterlogging management module combines short-term precipitation forecast and real-time hydrological data, uses advanced information technology and hydrological models to predict waterlogging risks, and takes corresponding preventive measures; specifically comprising the following steps:

[0037] Step C1, data collection and monitoring: deploy intelligent sensing terminal equipment to collect urban waterlogging-related hydrological data in real time and transmit it to the central processing system;

[0038] Hydrological data flow calculation: Q = λPAT; where Q is the design flow, λ is the runoff coefficient, P is the rainfall intensity, A is the catchment area, and T is the rainfall duration;

[0039] Step C2, flood simulation and risk assessment: Use hydrological models and flood simulation technology, combined with real-time hydrological data and short-term precipitation forecasts, to assess waterlogging risks;

[0040] The specific calculation of waterlogging risk assessment is as follows: R = f(Q, C, P); where R is the waterlogging risk level, Q is the design flow, C is the allowable concentration of the river discharge, and P is the pollutant monitoring concentration of rainwater;

[0041] Step C3, early warning and emergency response: Based on the risk assessment results, early warning information is promptly issued through the intelligent early warning system, and corresponding emergency response measures are initiated;

[0042] Step C4, linkage control: Linking with the urban drainage system, intelligently regulating drainage facilities, including drainage pumping stations, to mitigate the impact of waterlogging.

[0043] As a further preferred solution of the coordinated four-pre-control system of non-point source pollution and waterlogging based on short-term precipitation forecast of the present invention, the four-pre-control platform is a comprehensive platform integrating forecast, early warning, rehearsal and emergency plan functions, aiming to achieve coordinated control of short-term precipitation, non-point source pollution and waterlogging; specifically comprising the following steps:

[0044] Step D1, data collection and processing: collect real-time data including water level, flow, and water quality through monitoring equipment, and perform data preprocessing and quality control;

[0045] Step D2, model operation and simulation: run the hydrological model and water quality model to simulate flood evolution and pollutant diffusion under different rainfall scenarios;

[0046] Among them, the rainfall-runoff model is specifically calculated as follows: Q = μA·P; where Q is the runoff, μA is the flow per unit area, and P is the rainfall;

[0047] The flood evolution model is calculated as follows: H = f(Q, A, S); where H is the flood level, Q is the flow, A is the cross-sectional area, and S is the slope;

[0048] Step D3, risk assessment and early warning: Based on the model results, assess the risks of waterlogging and non-point source pollution, issue early warning information, and initiate emergency response measures;

[0049] Step D4, plan formulation and optimization: formulate and optimize response plans based on simulation results and risk assessment, including scheduling plans, material deployment, and personnel evacuation;

[0050] Step D5, decision support and emergency response: provide decision support, including automatic generation of scheduling plans, manual optimization and linkage rehearsal, and optimally propose the best scheduling plan.

[0051] As a further preferred solution of the non-point source pollution and waterlogging coordinated four-pre-control system based on short-term precipitation forecast of the present invention, the intelligent decision support system uses data analysis technology to assist decision-making; in urban water environment management, IDSS can provide decision support based on data analysis, including emergency response, resource allocation and pollution control strategies; specifically includes the following steps;

[0052] Step E1, data integration: collect and integrate meteorological, hydrological and environmental data to provide comprehensive data support for decision-making;

[0053] Step E2, model building: building machine learning models and optimization models for prediction and optimization of decision-making solutions;

[0054] Step E3, simulation: evaluate the effects of different decision-making schemes through simulation technology;

[0055] Step E4, decision support: provide decision support, including emergency response plans, resource allocation plans and pollution control strategies;

[0056] Among them, the resource allocation optimization model is specifically calculated as follows:

[0057] Minimize

[0058] Subject to:

[0059]

[0060] Among them, xij is the resource allocation volume from resource point i to demand point j, c ij is the transportation cost per unit of resource, a ij is the consumption rate of resources, b i is the resource volume of resource point i, d j is the demand volume of demand point j;

[0061] Pollution control strategy optimization model, the specific calculation is as follows:

[0062] Minimize

[0063] Subject to:

[0064] y k ∈{0,1};

[0065] wherein, wherein, y k is the binary variable indicating whether the k-th pollution control measure is adopted, p k is the cost of measure k, e kj is the emission reduction effect of measure k on pollution source j, E j is the allowable emission volume of pollution source j;

[0066] Step E5, human-computer interaction: Design an intuitive user interface to enable decision-makers to interact with the system, adjust parameters and plans.

[0067] As a further preferred solution of the present invention for a short-term precipitation forecast-based collaborative four-prevention and control system for non-point source pollution and waterlogging, the real-time monitoring and feedback system monitors water quality, water level and flow in real time by deploying a sensor network to ensure real-time update and feedback of data; specifically includes the following;

[0068] Sensor network deployment: Dynamically adjust the sensor positions according to environmental changes;

[0069] Water quality, water level, and flow monitoring: Water quality monitoring uses sensors that can detect multiple water quality indicators, including dissolved oxygen sensors, turbidity sensors, pH value sensors, and conductivity sensors; water level monitoring selects pressure-type water level sensors and ultrasonic water level sensors; flow monitoring selects appropriate flow meters according to the characteristics of the reservoir and the water flow situation, including ultrasonic flow meters and electromagnetic flow meters;

[0070] Real-time data update and feedback mechanism: The data acquisition equipment uses a data collector to collect and organize the data detected by the sensor, and transmits the data to the monitoring center through wired methods including Ethernet, RS485 or wireless methods including GPRS, LoRa, and NB-IoT; the server and database of the monitoring center are used to store and process the collected data, and the monitoring software has data display, analysis, and alarm functions, which presents the real-time and historical data of water quality, water level and flow in an intuitive way; by establishing a data analysis model, the collected data is analyzed, the flow is estimated through the water level-flow relationship curve, and the trend of water quality changes is judged.

[0071] Compared with the prior art, the present invention adopts the above technical solution and has the following technical effects:

[0072] 1. The present invention provides a coordinated four-pre-control system for non-point source pollution and waterlogging based on short-term precipitation forecasts. By integrating meteorological monitoring, hydrological monitoring and environmental monitoring data, it uses machine learning algorithms to achieve high-precision short-term precipitation forecasts, and dynamically adjusts the operation strategies of intelligent interception wells and initial rain storage pools to control non-point source pollution. At the same time, it combines real-time hydrological data to predict waterlogging risks and take preventive measures. This technology realizes the integrated management of forecasts, warnings, rehearsals and plans through the four-pre-control platform, supplemented by an intelligent decision support system and a real-time monitoring and feedback system, as well as a water ecological governance module under the sponge city design concept, to comprehensively improve the efficiency and effectiveness of urban water environment management;

[0073] 2. The present invention can realize high-precision forecast of upcoming precipitation events by integrating the short-term precipitation forecast module, thereby improving the city's response speed and coping ability to extreme weather events; the non-point source pollution control module can dynamically adjust the operation strategy of the intelligent interception well and the initial rain storage pool according to the short-term precipitation forecast, effectively reduce the discharge of non-point source pollutants, and protect the water body from pollution; the waterlogging management module combines the short-term precipitation forecast and real-time hydrological data to predict the waterlogging risk and take corresponding preventive measures, thereby significantly improving the city's waterlogging defense and management capabilities; the four-pre-control platform integrates the forecast, early warning, rehearsal, and plan functions to realize the coordinated control of short-term precipitation, non-point source pollution, and waterlogging, thereby improving the efficiency and effectiveness of urban water environment management; the intelligent decision support system provides decision support based on data analysis, including emergency response, resource Source allocation and pollution control strategies help decision makers formulate more scientific and reasonable management measures; improve monitoring efficiency and data feedback speed: the real-time monitoring and feedback system deploys a sensor network to monitor water quality, water level and flow in real time, ensure real-time data update and feedback, and improve monitoring efficiency and data feedback speed; the water ecological governance module integrates the sponge city design concept, improves water environment quality through ecological restoration and water purification technology, and promotes the healthy development of urban water ecosystems; through comprehensive management and control of urban water environment, enhances the city's adaptability and resilience to climate change and extreme weather events; by improving the quality of water environment, it helps to achieve sustainable use of water resources and support the sustainable development of cities; through effective water environment management and disaster prevention, it can improve the public's satisfaction and sense of security with urban water environment management. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without paying any creative work.

[0075] Figure 1 is a system architecture diagram of the present invention;

[0076] Figure 2 is a flow chart of a non-point source pollution control module of the present invention;

[0077] Figure 3 is a flow chart of a waterlogging management module of the present invention;

[0078] Figure 4 It is a flow chart of the four pre-control platforms of the present invention;

[0079] Figure 5 is a flow chart of the intelligent decision support system of the present invention;

[0080] Figure 6 is a schematic diagram of the real-time monitoring and feedback system of the present invention;

[0081] Figure 7 It is a schematic diagram of the water ecological management module of the present invention. DETAILED DESCRIPTION

[0082] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings:

[0083] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. The present invention is described in detail below based on the drawings and preferred embodiments, and the purpose and effect of the present invention will become clearer. It should be understood that the specific embodiments described here are only used to explain the present invention and are not used to limit the present invention.

[0084] The present invention relates to the field of urban water environment management technology, in particular to a non-point source pollution and waterlogging coordinated four-prediction (forecast, warning, rehearsal, and plan) control system based on short-term precipitation forecast. This technical field covers the following key directions:

[0085] Short-term precipitation forecasting technology: This technical field includes methods for short-term precipitation forecasting using advanced technologies such as deep learning and dual-polarization radar, which are of great significance for daily protection of life and property. The development of short-term precipitation forecasting technology, especially the application of deep learning methods in forecasting, provides new opportunities for improving short-term forecasting of convective storms.

[0086] Urban Storm Waterlogging Management Technology: This field focuses on the characteristics, mechanisms, data and methods of urban storm waterlogging, and how to use new technologies in the fields of earth science and computer science, such as multi-source remote sensing technology, weather radar, social media and crowdsourcing data, and artificial intelligence technology, to conduct urban hydrological research and storm waterlogging simulation.

[0087] Non-point source pollution control technology: involves a variety of technologies for the prevention and control of farmland non-point source pollution, including soil tillage optimization, application of soil conditioners, ammonia volatilization control technology, water-saving irrigation technology, farmland waste treatment technology, pesticide reduction and residue control technology, as well as farmland non-point source pollution process blocking technology and terminal enhancement technology, such as ecological ridge technology, ecological interception belt technology, ecological ditch interception technology, pre-storage technology and artificial wetland technology.

[0088] Four-pre-control technology: An important part of the construction of intelligent water conservancy, which involves four links: forecasting, early warning, pre-simulation, and pre-plan. It is the starting point and ending point of the construction of the digital twin basin and also the main standard for testing the achievements of the digital twin basin construction. The four-pre functions aim to achieve comprehensive forecasting, early warning, pre-simulation, and pre-plan, and improve the water regime monitoring and intelligent dispatching capabilities.

[0089] Intelligent decision support system: Combining artificial intelligence and expert system technologies, enabling the decision support system to make more full use of human knowledge and helping to solve complex decision-making problems through logical reasoning. The application of the intelligent decision support system in urban water environment management helps to achieve fine and comprehensive management and control of the water environment.

[0090] Real-time monitoring and feedback system: In the DevOps culture, continuous monitoring and feedback are the key links to ensure the reliability and stability of the system. Through effective monitoring and timely feedback mechanisms, the team can quickly discover and solve potential problems, thereby improving the availability of the system and user satisfaction.

[0091] Water ecological governance technology: Including river and lake water pollution ecological restoration technologies, such as biofilm restoration technology, constructed wetland technology, ecological floating island technology, immobilized bio-enzyme technology, and aeration and oxygenation technology. These technologies use ecological and biological methods to transfer, transform, and degrade pollutants in water to achieve water purification.

[0092] A collaborative four-pre-control system for non-point source pollution and waterlogging based on short-term and imminent precipitation forecasting, as Figure 1 shown, includes a short-term and imminent precipitation forecasting module, a non-point source pollution control module, a waterlogging management module, a four-pre-control platform, an intelligent decision support system, a real-time monitoring and feedback system, and a water ecological governance module;

[0093] Among them, the short-term and imminent precipitation forecasting module is used to achieve high-precision forecasting of short-term and imminent precipitation by using meteorological data and machine learning algorithms;

[0094] The non-point source pollution control module is used to adjust the operation strategies of intelligent intercepting wells and first-flush storage ponds according to the short-term and imminent precipitation forecasting to control non-point source pollution;

[0095] The waterlogging management module is used to combine the short-term and imminent precipitation forecasting and real-time hydrological data to predict the waterlogging risk and take corresponding preventive measures;

[0096] The four-pre-control platform is used to integrate the functions of forecasting, early warning, pre-simulation, and pre-plan to achieve collaborative management and control of short-term and imminent precipitation, non-point source pollution, and waterlogging;

[0097] The intelligent decision support system is used to provide decision support based on data analysis, including emergency response, resource allocation, and pollution control strategies;

[0098] Real-time monitoring and feedback system for deploying a sensor network to monitor water quality, water level, and flow rate in real time, ensuring real-time data update and feedback;

[0099] Water ecological governance module for integrating the design concept of sponge city and improving water environment quality through ecological restoration and water purification technologies.

[0100] The short-term and imminent precipitation forecasting module mainly realizes high-precision forecasting of future precipitation based on deep learning models and dual-polarization radar data. This module uses deep learning models, especially convolutional neural networks (CNNs) and generative adversarial networks (GANs), to process and analyze meteorological data for predicting short-term and imminent precipitation events. Dual-polarization radar can provide information such as the size, phase state, and water content of precipitation particles, which is crucial for the identification of precipitation types and the estimation of precipitation intensity.

[0101] Specifically, it includes the following steps;

[0102] Step A1, data preprocessing: Read the horizontal reflectivity factor Z in the dataset h data, design different threshold segmentation algorithms, and eliminate precipitation events with less or no precipitation in the dataset

[0103] to obtain data meeting the requirements of heavy precipitation;

[0104] Step A2, establish a deep learning model: Use deep learning models, including ConvLSTM, RNN, and Transformer, to directly predict the future rainfall rate using radar; the model needs to be able to handle complex non-linear spatio-temporal conversions to reduce prediction ambiguity; the specific calculation is as follows:

[0105] Horizontal reflectivity factor where D is the diameter of the particle, N(D) is the number concentration of particles per unit volume of air and per unit size, λ is the radar wavelength, D m is the maximum and minimum values of the particle diameter, K w is the dielectric constant of the particle, f h (π, D) is the backscattering amplitude in the horizontal polarization direction;

[0106] Differential reflectivity where Z v is the vertical reflectivity factor;

[0107] Specific differential phase shift where R e is the real part value of the integral part, f h (0, D) and f v (0, D) are the forward scattering amplitudes in the horizontal polarization direction and the vertical polarization direction respectively;

[0108] Step A3, Model Accuracy Evaluation: By quantitatively evaluating the prediction accuracy of the model, including comparing with the optical flow method, traditional deep learning methods, and the U-Net method, evaluate the forecasting effect of the model on different time scales.

[0109] As Figure 2 shown, the non-point source pollution control module controls non-point source pollution through the coordinated operation of intelligent intercepting wells and first flush storage ponds; using short-term precipitation forecast data, dynamically adjusts the operation strategies of these two facilities to reduce pollutants from entering the water body;

[0110] The non-point source pollution control module mainly controls non-point source pollution through the coordinated operation of intelligent intercepting wells and first flush storage ponds. This module uses short-term precipitation forecast data to dynamically adjust the operation strategies of these two facilities to reduce pollutants from entering the water body.

[0111] Intelligent Intercepting Well: The intelligent intercepting well realizes the control of the inflow through its hydraulic control system and monitoring system (including pressure sensors and rain gauges). Under the guidance of precipitation forecasts, the intelligent intercepting well can adjust the opening and closing states of the pollution interception gate and the rainwater gate in advance according to the predicted precipitation amount and intensity to intercept pollutants.

[0112] First Flush Storage Pond: The first flush storage pond is used to collect the initial rainwater, which contains a high concentration of pollutants. Through short-term precipitation forecasts, the storage pond can make preparations in advance and adjust its operation strategy, such as emptying in advance or increasing the storage capacity, to cope with the upcoming precipitation event.

[0113] Specifically, it includes the following steps;

[0114] Step B1, Storage Pond Volume Calculation: According to short-term rainfall forecasts, determine the rainfall duration and rainfall intensity; use the terrain and land use conditions to determine the catchment area and runoff coefficient; calculate the required volume of the storage pond to ensure effective control of non-point source pollution during rainfall;

[0115] Storage Pond Volume Calculation: Considering factors such as short-term rainfall, surface scouring, and sediment, the calculation formula for the storage pond volume is expressed as: Where: V is the effective volume of the storage pond; F is the catchment area; ψ is the runoff coefficient, considering the influence of rainfall and ground conditions; Δt is the rainfall duration, determined according to short-term rainfall forecasts; r is the rainfall intensity, also determined according to short-term rainfall forecasts; ψz is the comprehensive rainfall runoff coefficient of the construction site;

[0116] Step B2, Pollutant Concentration Calculation: Determine the initial pollutant concentration at the start of rainfall through monitoring; calculate the change in pollutant concentration at different time points according to the rainfall intensity and rainfall duration; adjust the operation strategy of the sewage treatment plant to adapt to the change in pollutant concentration under different rainfall conditions;

[0117] Calculation of pollutant concentration: Considering factors such as the allowable amount of initial rainwater discharged to the sewage treatment plant, the monitored concentration of pollutants in rainwater, the allowable concentration for discharging into the river, and the rainfall duration, the calculation formula for pollutant concentration is expressed as:

[0118] Where: C t is the pollutant concentration at a certain moment; C0 is the initial pollutant concentration at the start of rainfall; k is the flushing coefficient, which is related to the ground flushing and sediment removal rate; r is the rainfall intensity; t is the rainfall duration; C bg is the background concentration, that is, the pollutant concentration during non-rainfall periods;

[0119] Step B3, adjustment of the intelligent intercepting well operation strategy: According to the short-term and imminent precipitation forecast, the control system of the intelligent intercepting well will adjust the opening and closing degrees of the sewage interception gate and the rainwater gate to control the inflow of pollutants. The specific strategy includes closing the sewage interception gate before predicting a large amount of precipitation to intercept pollutants, and at the same time opening the rainwater gate to reduce the risk of waterlogging;

[0120] Step B4, adjustment of the initial rainwater storage tank operation strategy: The initial rainwater storage tank will adjust its operation mode according to the precipitation forecast. Before predicting a precipitation event, the storage tank can be emptied in advance to reserve storage space for the upcoming rainwater. During precipitation, the storage tank will collect the initial rainwater and treat it after the rain to reduce pollutant emissions.

[0121] As Figure 3 shown, the waterlogging management module combines short-term and imminent precipitation forecasts and real-time hydrological data, and through advanced information technology and hydrological models, predicts the waterlogging risk and takes corresponding preventive measures;

[0122] The core of the waterlogging management module lies in combining short-term and imminent precipitation forecasts and real-time hydrological data, and through advanced information technology and hydrological models, predicting the waterlogging risk and taking corresponding preventive measures. The technical principles of this module include the following aspects:

[0123] Short-term and imminent precipitation forecasting technology: Using a multi-scale precipitation dynamic monitoring system constructed by a meteorological radar observation system and ground rain gauges, combined with quantitative precipitation estimation (QPE) technology, to improve the refinement level of precipitation forecasting and the lead time of heavy precipitation.

[0124] Real-time hydrological data monitoring: Through intelligent sensing terminal devices, including water level gauges, flow meters, etc., real-time monitoring of key hydrological data of urban waterlogging, including rainfall, groundwater level, river water level, etc.

[0125] Flood process simulation: Adopting a method of dual-driving by mechanism model and intelligent model and iterative update of self-learning mode to improve the accuracy and operation efficiency of flood simulation.

[0126] Risk assessment and early warning: Based on precipitation forecasts and flood simulation results, analytical techniques such as decision tree models are used to achieve early identification and early warning of flood risks.

[0127] The specific steps include:

[0128] Step C1, data collection and monitoring: deploy intelligent sensing terminal equipment to collect urban waterlogging-related hydrological data in real time and transmit it to the central processing system;

[0129] Hydrological data flow calculation: Q = λPAT; where Q is the design flow, λ is the runoff coefficient, P is the rainfall intensity, A is the catchment area, and T is the rainfall duration;

[0130] Step C2, flood simulation and risk assessment: Use hydrological models and flood simulation technology, combined with real-time hydrological data and short-term precipitation forecasts, to assess waterlogging risks;

[0131] The specific calculation of waterlogging risk assessment is as follows: R = f(Q, C, P); where R is the waterlogging risk level, Q is the design flow, C is the allowable concentration of the river discharge, and P is the pollutant monitoring concentration of rainwater;

[0132] Step C3, early warning and emergency response: Based on the risk assessment results, early warning information is promptly issued through the intelligent early warning system, and corresponding emergency response measures are initiated;

[0133] Step C4, linkage control: Linking with the urban drainage system, intelligently regulating drainage facilities, including drainage pumping stations, to mitigate the impact of waterlogging.

[0134] like Figure 4 As shown, the four-prediction control platform is a comprehensive platform that integrates forecasting, warning, rehearsal, and emergency plan functions, aiming to achieve coordinated control of short-term precipitation, non-point source pollution, and waterlogging. The technical principles of the platform involve the following aspects:

[0135] Data integration and analysis: The platform collects data from multiple sources such as meteorology, hydrology, and environment, and uses big data analysis technology to integrate information and conduct in-depth analysis.

[0136] Model-driven prediction and simulation: Use hydrological models, water resource models, water quality models, etc. to simulate and predict the impact of different decision-making options on the water conservancy system.

[0137] Intelligent early warning system: Based on model algorithms, it can realize early warning of disasters such as floods and urban waterlogging. By broadening the channels for issuing early warning information, the early warning information can be directly delivered to the public in the affected areas in a timely manner.

[0138] Digital twin technology: Use digital twin technology to digitally model and simulate watersheds and water conservancy projects, and realize digital mapping and intelligent simulation of all elements of the physical watershed and the entire process of water conservancy management.

[0139] Multi-objective optimization: In terms of non-point source pollution control, multi-objective collaborative optimization technology is used to achieve optimal configuration of best management measures (BMPs);

[0140] The specific steps include:

[0141] Step D1, data collection and processing: collect real-time data including water level, flow, and water quality through monitoring equipment, and perform data preprocessing and quality control;

[0142] Step D2, model operation and simulation: run the hydrological model and water quality model to simulate flood evolution and pollutant diffusion under different rainfall scenarios;

[0143] Among them, the rainfall-runoff model is specifically calculated as follows: Q = μA·P; where Q is the runoff, μA is the flow per unit area, and P is the rainfall;

[0144] The flood evolution model is calculated as follows: H = f(Q, A, S); where H is the flood level, Q is the flow, A is the cross-sectional area, and S is the slope;

[0145] Step D3, risk assessment and early warning: Based on the model results, assess the risks of waterlogging and non-point source pollution, issue early warning information, and initiate emergency response measures;

[0146] Step D4, plan formulation and optimization: formulate and optimize response plans based on simulation results and risk assessment, including scheduling plans, material deployment, and personnel evacuation;

[0147] Step D5, decision support and emergency response: provide decision support, including automatic generation of scheduling plans, manual optimization and linkage rehearsal, and optimally propose the best scheduling plan.

[0148] like Figure 5 As shown, the intelligent decision support system (IDSS) is a system that uses data analysis technology to assist decision making. In urban water environment management, IDSS can provide decision support based on data analysis, including emergency response, resource allocation and pollution control strategies. The following are the key technical principles of the system:

[0149] Data fusion and analysis: Integrate and fuse data from different sources, including meteorological data, hydrological data, environmental monitoring data, etc., for comprehensive analysis.

[0150] Machine Learning and Prediction Models: Apply machine learning algorithms, including random forests, neural networks, etc., to perform pattern recognition and predictive analysis on data.

[0151] Optimization Algorithms: Use optimization algorithms in operations research, including linear programming, non-linear programming, multi-objective optimization, etc., to formulate the best decision-making solutions.

[0152] Simulation and Emulation: Through simulation and emulation technologies, preview the effects of different decision-making solutions and evaluate their impacts on the environment and socio-economy.

[0153] Human-Computer Interaction: Provide an intuitive user interface and interaction design, enabling decision-makers to easily input parameters, view results, and adjust solutions.

[0154] Specifically, it includes the following steps;

[0155] Step E1, Data Integration: Collect and integrate meteorological, hydrological, and environmental data to provide comprehensive data support for decision-making;

[0156] Step E2, Model Construction: Construct machine learning models and optimization models for predicting and optimizing decision-making solutions;

[0157] Step E3, Simulation and Emulation: Through simulation and emulation technologies, evaluate the effects of different decision-making solutions;

[0158] Step E4, Decision Support: Provide decision support, including emergency response plans, resource allocation plans, and pollution control strategies;

[0159] Among them, the resource allocation optimization model is specifically calculated as follows:

[0160] Minimize

[0161] Subject to:

[0162]

[0163] Among them, x ij is the resource allocation quantity from resource point i to demand point j, c ij is the transportation cost per unit of resource, a ij is the consumption rate of the resource, b i is the resource quantity at resource point i, d j is the demand quantity at demand point j;

[0164] The pollution control strategy optimization model is specifically calculated as follows:

[0165] Minimize

[0166] Subject to:

[0167] y k ∈ {0, 1};

[0168] where y k is a binary variable indicating whether the k-th pollution control measure is adopted, p k is the cost of measure k, e kj is the emission reduction effect of measure k on pollution source j, E j is the allowable emission of pollution source j;

[0169] Step E5, Human-computer Interaction: Design an intuitive user interface to enable decision-makers to interact with the system, adjust parameters and plans.

[0170] As Figure 6 shown, the real-time monitoring and feedback system: Deploy a sensor network to monitor water quality, water level and flow rate in real time, ensuring real-time update and feedback of data.

[0171] The real-time monitoring and feedback system deploys a sensor network to monitor water quality, water level and flow rate in real time, ensuring real-time update and feedback of data. The core of this system lies in the real-time data feedback technology, which usually uses polling or long connections via WebSocket for real-time communication. On the Node.js side, NPM packages such as http-event-stream can be installed for server-side push to achieve the function of "pushing" server data to the client in real time.

[0172] Deployment of Sensor Network: The deployment of the sensor network determines the coverage of the network and the quality of data collection. Common deployment methods include uniform deployment, cluster deployment and dynamic deployment. The uniform deployment method provides full coverage but may lead to energy waste. The cluster deployment method reduces the communication volume between nodes, reduces energy consumption and improves data transmission efficiency. The dynamic deployment method adjusts the positions of sensors dynamically according to environmental changes.

[0173] Monitoring of Water Quality, Water Level and Flow Rate: Water quality monitoring usually uses sensors that can detect multiple water quality indicators, including dissolved oxygen sensors, turbidity sensors, pH value sensors, conductivity sensors, etc. For water level monitoring, pressure water level sensors, ultrasonic water level sensors, etc. can be selected. For flow rate monitoring, appropriate flow meters are selected according to the characteristics of the reservoir and the water flow situation, including ultrasonic flow meters, electromagnetic flow meters, etc.

[0174] Data real-time update and feedback mechanism: The data acquisition device uses a data collector to collect and organize the data detected by the sensors, and transmits the data to the monitoring center via wired (including Ethernet, RS485, etc.) or wireless (including GPRS, LoRa, NB-IoT, etc.) means. The servers and databases in the monitoring center are used to store and process the collected data. The monitoring software has functions such as data display, analysis, and alarm, presenting the real-time situation and historical data of water quality, water level, and flow rate in an intuitive manner. In addition, by establishing a data analysis model, the collected data is analyzed, for example, the flow rate is calculated through the water level-flow rate relationship curve, and the trend of water quality change is judged.

[0175] Specifically, it includes the following:

[0176] Sensor network deployment: Dynamically adjust the sensor positions according to environmental changes;

[0177] Water quality, water level, and flow rate monitoring: For water quality monitoring, sensors that can detect multiple water quality indicators are used, including dissolved oxygen sensors, turbidity sensors, pH value sensors, and conductivity sensors; for water level monitoring, pressure water level sensors and ultrasonic water level sensors are selected; for flow rate monitoring, according to the characteristics of the reservoir and the water flow situation, appropriate flow meters are selected, including ultrasonic flow meters and electromagnetic flow meters;

[0178] Data real-time update and feedback mechanism: The data acquisition device uses a data collector to collect and organize the data detected by the sensors, and transmits the data to the monitoring center via wired including Ethernet, RS485 or wireless including GPRS, LoRa, NB-IoT; the servers and databases in the monitoring center are used to store and process the collected data, and the monitoring software has data display, analysis, and alarm functions, presenting the real-time situation and historical data of water quality, water level, and flow rate in an intuitive manner; by establishing a data analysis model, the collected data is analyzed, and the flow rate is calculated through the water level-flow rate relationship curve, and the trend of water quality change is judged.

[0179] As Figure 7 shown, the water ecological governance module integrates the design concept of sponge city, and improves the water environment quality through ecological restoration and water purification technologies.

[0180] (1) Integrating the design concept of sponge city:

[0181] The core of the sponge city design concept lies in realizing the natural accumulation, natural infiltration, and natural purification of water, and promoting the formation of an ecological, safe, and sustainable urban water cycle system. This concept protects and utilizes ecological spaces such as natural mountains, rivers, lakes, wetlands, cultivated land, forests, and grasslands in the city, gives play to the rainwater absorption and slow release effects of buildings, roads, green spaces, and water systems, and improves the city's water storage, water infiltration, and water conservation capabilities.

[0182] (2) Ecological restoration technologies:

[0183] ① Biofilm restoration technology: Using natural materials or synthetic contact materials as carriers, enabling microbial populations to adhere to the carrier surface in a membranous form. Through contact with sewage, the microorganisms on the biofilm take in the organic matter in the sewage as nutrients and assimilate it, thereby purifying the sewage.

[0184] ② Constructed wetland technology: The restoration of natural wetlands and the construction of constructed wetlands are important engineering measures for controlling pollution at the estuaries of lakes entering the lake at present, and it is an economical, effective, feasible method with good ecological functions.

[0185] ③ Ecological floating island technology: Using floating materials as substrates or carriers, planting higher aquatic plants or terrestrial plants into eutrophic waters. Through the absorption or adsorption of the plant roots, nitrogen, phosphorus, and organic pollutants in the water body are reduced, thereby purifying the water quality.

[0186] ④ Aeration and oxygenation technology: A method to increase the oxygen content in water, which is a commonly used technology in the treatment of river and lake pollution to increase the dissolved oxygen content and reduce the COD to meet the standard.

[0187] (3) Water purification technologies:

[0188] ① Water quality ecological purification technology: Centered on the construction of wetland plants or aquatic plant communities, a series of technologies that use plants themselves and their symbiotic biological systems to remove pollutants in the water body. It includes constructed wetlands, biological floating beds, artificial sinking beds, underwater forests, riparian buffer zones, etc.

[0189] ② Constructed wetland water purification technology: Utilizing the self-regulation mechanism of the soil-microorganism-plant ecosystem and its comprehensive purification function for pollutants to improve the water quality to varying degrees.

[0190] ③ Ecological interception technology: On the premise of intercepting and controlling pollution sources, using riverine zones and lakeshore zones to intercept and reduce runoff pollution or non-point source pollution to reduce the pollution load entering rivers (lakes).

[0191] (4) Improvement of water environment quality:

[0192] Through the above ecological restoration and water purification technologies, the water ecological governance module can effectively improve the water environment quality. These technologies can not only reduce the pollutants in the water body, but also restore and enhance the self-purification ability of the water body, promote aquatic biodiversity, thereby realizing the systematic governance and coordinated promotion of water resources, water environment, and water ecology.

[0193] Example: Synergistic four-prevention control technology for non-point source pollution and waterlogging in Jianye District, Nanjing City

[0194] 1. Background and objectives:

[0195] As an important area in the lower reaches of the Yangtze River, Jianye District of Nanjing faces problems such as interweaving water pollution risks, imbalance of river and lake ecosystems, and pressure on flood control in the embankment area. This embodiment aims to improve the water environment quality of Jianye District, effectively control non-point source pollution, and enhance waterlogging management capabilities by applying this patented technology.

[0196] 2. Short-term precipitation forecast module:

[0197] (1) Data collection: Use monitoring equipment in the smart water system, including water level sensors, rain gauges, and weather stations, to collect real-time hydrological and meteorological data.

[0198] (2) Model training and real-time forecasting: Based on historical flood events and related data, a flood prediction model is established to generate flood risk forecasts in real time.

[0199] 3. Non-point source pollution control module:

[0200] (1) Intelligent interception well adjustment: According to the short-term precipitation forecast, the operation strategy of the intelligent interception well is automatically adjusted, such as closing the sewage interception gate in advance to intercept pollutants.

[0201] (2) Management of initial rainwater storage ponds: Adjust the storage capacity of initial rainwater storage ponds to collect and process initial rainwater and reduce pollutant emissions.

[0202] 4. Waterlogging management module:

[0203] (1) Hydrological data monitoring: Deploy sensors such as water level meters and flow meters to monitor key hydrological data of urban waterlogging in real time.

[0204] (2) Flood simulation and risk assessment: Use hydrological models to simulate flood evolution under different rainfall scenarios and assess the risk of waterlogging.

[0205] 5. Four pre-control platforms:

[0206] (1) Integrated system: Integrates forecasting, warning, rehearsal, and emergency planning functions to achieve coordinated control of short-term precipitation, non-point source pollution, and urban waterlogging.

[0207] (2) Plan formulation: Based on the risk assessment results, formulate and optimize response plans, including scheduling plans, material allocation, and personnel evacuation.

[0208] 6. Intelligent decision support system:

[0209] (1) Data analysis: Collect and analyze meteorological, hydrological, environmental and other data to provide support for decision-making.

[0210] (2) Model building and decision support: Build machine learning models and optimization models to predict and optimize decision solutions.

[0211] 7. Real-time Monitoring and Feedback System:

[0212] (1) Sensor Network Deployment: Deploy a sensor network to monitor water quality, water level, and flow rate in real time.

[0213] (2) Real-time Data Update and Feedback: Transmit data to the monitoring center via wired or wireless means to achieve real-time data update.

[0214] 8. Water Ecosystem Governance Module:

[0215] (1) Sponge City Design: Integrate the concept of Sponge City design and improve the water environment quality through ecological restoration and water purification technologies.

[0216] (2) Application of Ecological Restoration Technologies: Apply biofilm restoration technology, constructed wetland technology, etc. to reduce pollutants in water bodies.

[0217] 9. Implementation Effects:

[0218] (1) Improve the Accuracy of Flood Forecasting: Improve the forecasting accuracy, extend the lead time, and provide a scientific basis for flood prevention and mitigation.

[0219] (2) Effectively Control Non-point Source Pollution: Through the coordinated operation of intelligent intercepting wells and first-flush storage ponds, reduce the discharge of non-point source pollutants.

[0220] (3) Enhance the Ability to Manage Waterlogging: Combine short-term precipitation forecasts and real-time hydrological data to predict waterlogging risks and take corresponding preventive measures.

[0221] (4) Improve the Water Environment Quality: Through ecological restoration and water purification technologies, improve the water quality of rivers and construct a healthy ecosystem.

[0222] The present invention can realize high-precision forecast of upcoming precipitation events by integrating the short-term precipitation forecast module, thereby improving the city's response speed and coping ability to extreme weather events; the non-point source pollution control module can dynamically adjust the operation strategy of the intelligent interception well and the initial rain storage pool according to the short-term precipitation forecast, effectively reduce the discharge of non-point source pollutants, and protect the water body from pollution; the waterlogging management module combines the short-term precipitation forecast and real-time hydrological data to predict the waterlogging risk and take corresponding preventive measures, thereby significantly improving the city's waterlogging defense and management capabilities; the four-pre-control platform integrates the forecast, early warning, rehearsal, and plan functions to realize the coordinated control of short-term precipitation, non-point source pollution, and waterlogging, thereby improving the efficiency and effectiveness of urban water environment management; the intelligent decision support system provides decision support based on data analysis, including emergency response, resource Allocation and pollution control strategies help decision makers formulate more scientific and reasonable management measures; improve monitoring efficiency and data feedback speed: the real-time monitoring and feedback system deploys a sensor network to monitor water quality, water level and flow in real time, ensure real-time data update and feedback, and improve monitoring efficiency and data feedback speed; the water ecological governance module integrates the sponge city design concept, improves water environment quality through ecological restoration and water purification technology, and promotes the healthy development of urban water ecosystems; through comprehensive management and control of urban water environment, enhances the city's adaptability and resilience to climate change and extreme weather events; by improving the quality of water environment, it helps to achieve sustainable use of water resources and support the sustainable development of cities; through effective water environment management and disaster prevention, it can improve the public's satisfaction and sense of security with urban water environment management.

[0223] Those skilled in the art can understand that the above are only preferred examples of the invention and are not intended to limit the invention. Although the invention is described in detail with reference to the above examples, those skilled in the art can still modify the technical solutions recorded in the above examples or replace some of the technical features with equivalents. Any modification, equivalent replacement, etc. made within the spirit and principle of the invention should be included in the protection scope of the invention. All technical features in this embodiment can be freely combined according to actual needs.

[0224] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A collaborative four-prevention control system for non-point source pollution and waterlogging based on short-term and imminent precipitation forecasting, characterized in that: It includes short-term precipitation forecast module, non-point source pollution control module, waterlogging management module, four-pre-control platform, intelligent decision support system, real-time monitoring and feedback system, and water ecological management module; Among them, the short-term precipitation forecast module is used to use meteorological data and machine learning algorithms to achieve high-precision forecasts of short-term precipitation; The non-point source pollution control module is used to adjust the operation strategy of the intelligent interception well and the initial rain storage pool according to the short-term precipitation forecast to control non-point source pollution; The waterlogging management module is used to combine short-term precipitation forecasts and real-time hydrological data to predict waterlogging risks and take corresponding preventive measures; Four pre-control platforms are used to integrate forecasting, warning, rehearsal and emergency plan functions to achieve coordinated control of short-term precipitation, non-point source pollution and waterlogging; Intelligent decision support system, used to provide decision support based on data analysis, including emergency response, resource allocation and pollution control strategies; Real-time monitoring and feedback system, which is used to deploy sensor networks to monitor water quality, water level and flow in real time, ensuring real-time data update and feedback; The water ecological management module is used to integrate the sponge city design concept and improve the water environment quality through ecological restoration and water purification technology.

2. The collaborative four-prediction control system for non-point source pollution and waterlogging based on short-term and imminent precipitation forecasting according to claim 1, wherein: The short-term precipitation forecast module realizes high-precision forecast of future precipitation based on deep learning model and dual-polarization radar data; uses deep learning model, especially convolutional neural network CNN and generative adversarial network GAN, to process and analyze meteorological data, so as to predict short-term precipitation events; specifically includes the following steps: Step A1, data preprocessing: Read the horizontal reflectivity factor Z in the dataset h data, design different threshold segmentation algorithms for the dataset Precipitation events with little or no precipitation are eliminated to obtain data that meets the requirements of heavy precipitation; Step A2, build a deep learning model: use deep learning models, including ConvLSTM, RNN, and Transformer, to directly predict future rainfall rates using radar; the model needs to be able to handle complex nonlinear spatiotemporal transformations to reduce the ambiguity of the prediction; the specific calculation is as follows: Horizontal reflectivity factor Z h : where D is the diameter of the particle, N(D) is the number concentration of particles per unit volume of air and per unit size, λ is the radar wavelength, D m is the maximum and minimum values of the particle diameter, K w is the dielectric constant of the particle, f h (π, D) is the backscattering amplitude in the horizontal polarization direction; Differential reflectivity Z ΔD : where Z v is the vertical reflectivity factor; Ratio differential phase shift K ΔD : Wherein, R e is the real part value of the integral part, f h (0, D) and f v (0, D) are the forward scattering amplitudes in the horizontal polarization direction and the vertical polarization direction, respectively; Step A3, model accuracy assessment: The prediction accuracy of the model is quantitatively evaluated, including comparisons using the optical flow method, traditional deep learning methods, and U-Net methods to evaluate the prediction effect of the model at different time scales.

3. The collaborative four-prevention and control system for non-point source pollution and waterlogging based on short-term and imminent precipitation forecasting according to claim 1, characterized in that: The non-point source pollution control module controls non-point source pollution through the coordinated operation of the intelligent interception well and the initial rain storage pool; uses the short-term precipitation forecast data to dynamically adjust the operation strategies of the two facilities to reduce the entry of pollutants into the water body; specifically includes the following steps: Step B1, calculation of the volume of the regulating reservoir: determine the duration and intensity of rainfall according to the short-term rainfall forecast; determine the catchment area and runoff coefficient using the topography and land use conditions; calculate the required volume of the regulating reservoir to ensure that non-point source pollution can be effectively controlled during rainfall; Calculation of the storage pond volume: Considering factors such as short-term rainfall, surface scouring, sediment, etc., the calculation formula for the storage pond volume is expressed as: Where: V is the effective volume of the storage pond; F is the catchment area; ψ is the runoff coefficient, considering the influence of rainfall and ground conditions; Δt is the rainfall duration, determined according to the short-term rainfall forecast; r is the rainfall intensity, also determined according to the short-term rainfall forecast; ψz is the comprehensive rainfall runoff coefficient of the construction site; Step B2, pollutant concentration calculation: determine the initial pollutant concentration at the beginning of rainfall through monitoring; calculate the change of pollutant concentration at different time points according to rainfall intensity and rainfall duration; adjust the operation strategy of the sewage treatment plant to adapt to the change of pollutant concentration under different rainfall conditions; Calculation of pollutant concentration: Taking into account the amount of initial rain allowed to be discharged to the sewage treatment plant, the pollutant monitoring concentration of rainwater, the allowable concentration for discharge to the river, the duration of rainfall and other factors, the calculation formula for pollutant concentration is expressed as follows: Where: C t is the pollutant concentration at a given time; C0 is the initial pollutant concentration at the start of rainfall; k is the wash-off coefficient, related to the ground wash-off and sediment removal rate; r is the rainfall intensity; t is the rainfall duration; C bg is the background concentration, i.e., the pollutant concentration during non-rainfall periods; Step B3, intelligent interception well operation strategy adjustment: According to the short-term precipitation forecast, the control system of the intelligent interception well will adjust the opening and closing degree of the sewage interception gate and the rainwater gate to control the inflow of pollutants. Specific strategies include closing the sewage interception gate before heavy precipitation is predicted to intercept pollutants, and opening the rainwater gate to reduce the risk of waterlogging; Step B4, initial rain storage tank operation strategy adjustment: The initial rain storage tank will adjust its operation mode according to the precipitation forecast. Before the precipitation event is predicted, the storage tank can be emptied in advance to leave storage space for the upcoming rainwater. During the precipitation period, the storage tank will collect the initial rainwater and process it after the rain to reduce the emission of pollutants.

4. A collaborative four-prevention control system for non-point source pollution and waterlogging based on short-term and imminent precipitation forecasting according to claim 1, characterized in that: The waterlogging management module combines short-term precipitation forecasts and real-time hydrological data, uses advanced information technology and hydrological models to predict waterlogging risks and take corresponding preventive measures; specifically, it includes the following steps: Step C1, data collection and monitoring: deploy intelligent sensing terminal equipment to collect urban waterlogging-related hydrological data in real time and transmit it to the central processing system; Hydrological data flow calculation: Q = λPAT; where Q is the design flow, λ is the runoff coefficient, P is the rainfall intensity, A is the catchment area, and T is the rainfall duration; Step C2, flood simulation and risk assessment: Use hydrological models and flood simulation technology, combined with real-time hydrological data and short-term precipitation forecasts, to assess waterlogging risks; The specific calculation of waterlogging risk assessment is as follows: R = f(Q, C, P); where R is the waterlogging risk level, Q is the design flow, C is the allowable concentration of the river discharge, and P is the pollutant monitoring concentration of rainwater; Step C3, early warning and emergency response: Based on the risk assessment results, early warning information is promptly issued through the intelligent early warning system, and corresponding emergency response measures are initiated; Step C4, linkage control: Linking with the urban drainage system, intelligently regulating drainage facilities, including drainage pumping stations, to mitigate the impact of waterlogging.

5. A collaborative four-prevention management and control system for non-point source pollution and waterlogging based on short-term and impending precipitation forecasting according to claim 1, characterized in that: The four-pre-control platform is a comprehensive platform integrating forecast, warning, rehearsal and emergency plan functions, aiming to achieve coordinated control of short-term precipitation, non-point source pollution and waterlogging; specifically, it includes the following steps: Step D1, data collection and processing: collect real-time data including water level, flow, and water quality through monitoring equipment, and perform data preprocessing and quality control; Step D2, model operation and simulation: run the hydrological model and water quality model to simulate flood evolution and pollutant diffusion under different rainfall scenarios; Among them, the rainfall-runoff model is specifically calculated as follows: Q = μA·P; where Q is the runoff, μA is the flow per unit area, and P is the rainfall; The flood evolution model is calculated as follows: H = f(Q, A, S); where H is the flood level, Q is the flow, A is the cross-sectional area, and S is the slope; Step D3, risk assessment and early warning: Based on the model results, assess the risks of waterlogging and non-point source pollution, issue early warning information, and initiate emergency response measures; Step D4, plan formulation and optimization: formulate and optimize response plans based on simulation results and risk assessment, including scheduling plans, material deployment, and personnel evacuation; Step D5, decision support and emergency response: provide decision support, including automatic generation of scheduling plans, manual optimization and linkage rehearsal, and optimally propose the best scheduling plan.

6. The collaborative four-prevention control system for non-point source pollution and waterlogging based on short-term and imminent precipitation forecasting according to claim 1, characterized in that: The intelligent decision support system is a system that uses data analysis technology to assist decision-making. In urban water environment management, IDSS can provide decision support based on data analysis, including emergency response, resource allocation and pollution control strategies. Specifically, it includes the following steps: Step E1, data integration: collect and integrate meteorological, hydrological and environmental data to provide comprehensive data support for decision-making; Step E2, model building: building machine learning models and optimization models for prediction and optimization of decision-making solutions; Step E3, simulation: evaluate the effects of different decision-making schemes through simulation technology; Step E4, decision support: provide decision support, including emergency response plans, resource allocation plans and pollution control strategies; Among them, the resource allocation optimization model is specifically calculated as follows: Among them, x ij is the resource allocation volume from resource point i to demand point j, c ij is the transportation cost per unit of resource, a ij is the consumption rate of the resource, b i is the resource volume of resource point i, d j is the demand volume of demand point j; Pollution control strategy optimization model, the specific calculation is as follows: where y k is a binary variable indicating whether the k-th pollution control measure is adopted, p k is the cost of measure k, e kj is the emission reduction effect of measure k on pollution source j, E j is the allowable emission of pollution source j; Step E5, Human-computer interaction: Design an intuitive user interface to enable decision makers to interact with the system and adjust parameters and scenarios.

7. A collaborative four-prevention control system for non-point source pollution and waterlogging based on short-term and impending precipitation forecasting according to claim 1, characterized in that: The real-time monitoring and feedback system deploys a sensor network to monitor water quality, water level and flow in real time to ensure real-time update and feedback of data; specifically includes the following: Sensor network deployment: dynamically adjust sensor locations based on environmental changes; Water quality, water level, and flow monitoring: Water quality monitoring uses sensors that can detect multiple water quality indicators, including dissolved oxygen sensors, turbidity sensors, pH sensors, and conductivity sensors; water level monitoring uses pressure water level sensors and ultrasonic water level sensors; flow monitoring uses appropriate flow meters based on the characteristics of the reservoir and water flow conditions, including ultrasonic flow meters and electromagnetic flow meters; Real-time data update and feedback mechanism: The data acquisition equipment uses a data collector to collect and organize the data detected by the sensor, and transmits the data to the monitoring center through wired methods including Ethernet, RS485 or wireless methods including GPRS, LoRa, and NB-IoT; the server and database of the monitoring center are used to store and process the collected data, and the monitoring software has data display, analysis, and alarm functions, which presents the real-time and historical data of water quality, water level and flow in an intuitive way; by establishing a data analysis model, the collected data is analyzed, the flow is estimated through the water level-flow relationship curve, and the trend of water quality changes is judged.

Images