A watershed water environment prediction and early warning method and system and a computer device

Abstract

The application discloses a kind of basin water environment prediction early warning method, comprising the following steps: S1: the hydrological water power mechanism prediction model of basin is built;S2: simulating and predicting the liquid level change of basin pipe network;S3: the numerical prediction model of basin is built;S4: predict the liquid level change data of basin pipe network;S5: the result generated by the hydrological water power mechanism prediction model, the numerical prediction model both respectively simulates and compares with actual liquid level change situation, and the risk of waterlogging, overflow and water quality of basin is predicted and early warned.It also discloses a kind of basin water environment prediction early warning system, comprising: basin all-element monitoring monitoring unit, basin prediction early warning unit and basin joint debugging joint control unit.The application realizes the intelligent management and control of basin water environment, promotes the multi-objective of basin, i.e.water safety target, water environment target and water ecological target management and fine management.

Description

Technical Field

[0001] This invention relates to the field of water environment technology, and more specifically to a method, system, and computer equipment for predicting and warning water environment in a watershed. Background Technology

[0002] In recent years, while there has been much research on the scheduling and management of water resources or water supply in river basins, there has been relatively little research on the comprehensive water environment management and scheduling related to river basin drainage systems. Furthermore, in river basin scheduling and management, smart water systems often suffer from problems such as low accuracy of predictive models, slow response speed, and unsatisfactory application results, making it difficult to effectively guide the multi-objective, refined management of river basins.

[0003] Because the boundary conditions of integrated watershed management are numerous and complex, refined watershed management relies on model predictions to guide scheduling. However, in existing technologies, the mechanistic models of smart water management platforms require continuous calibration based on a large amount of empirical data to achieve high accuracy, which is difficult to accomplish in the short term. Furthermore, mechanistic models have characteristics such as long preparation times and high data reliability requirements, making it difficult to consistently meet the requirements of rapid response in watershed scheduling. While empirical models for smart water management platforms are widely used due to their short construction time, fast response speed, and high reliability, existing empirical data cannot fully meet the complex operating conditions of watersheds.

[0004] Therefore, it is inevitable that empirical models and mechanistic models will work together to drive smart water management platforms to guide watershed scheduling and management. Summary of the Invention

[0005] The present invention provides a dual-model driven method, system and computer equipment for predicting and warning water environment in watersheds, which can at least solve one of the above-mentioned technical problems.

[0006] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: A method for predicting and warning about watershed water environment includes the following steps: S1: Based on existing watershed network data, combined with monitoring of liquid levels and flow rates at key points in the network, and considering the watershed's runoff patterns, and based on the operational logic between the watershed network, pumping stations, sewage treatment facilities, and rivers, the SWMM / MIKE / infoworks model is used to simulate the operational patterns of various drainage facilities in the watershed's "network-station-plant-river" system, and a watershed hydrological and hydrodynamic mechanism prediction model is constructed. S2: Before or during rainfall, based on rainfall forecast data and real-time monitoring data provided by the professional database of the watershed full-element monitoring and control unit, the established hydrological and hydrodynamic mechanism prediction model is used to simulate and predict the liquid level changes in the watershed pipe network. S3: Utilize machine learning algorithms to analyze historical rainfall event data, establish a functional relationship between the operating status of different water facilities and the changes in pipe network liquid level at key control points in the basin under different rainfall conditions, and construct a numerical prediction model; S4: Before or during rainfall, the numerical prediction model combines rainfall forecast data and real-time monitoring data, and uses the established functional relationship to predict changes in the watershed pipeline network liquid level; S5: Based on the hydrological and hydrodynamic mechanism prediction model and the numerical prediction model, the platform predicts the time of overflow level occurrence. Considering the safety and real-time responsiveness of the scheduling, the prediction results of the two models are compared, and the prediction result of the model that first shows the overflow is selected as a reference for the formulation of scheduling strategy. S6: Based on the combined prediction results of the hydrological and hydrodynamic mechanism prediction model and the numerical prediction model, and according to the scheduling rules formulated in the scheduling scheme, an optimization algorithm is used to find the best solution and a real-time control scheduling model is constructed. S7: Construct an experience-based scheduling model based on the scheduling experience database accumulated from previous rainfall events, including rainfall, pipeline liquid levels, and plant facility operation status. S8: Combining the real-time control and scheduling model with the experience scheduling model, a recommended scheduling strategy and scheduling instructions are given. The management personnel determine the final scheduling plan based on the rainfall and the operating status of the water conservancy facilities in the basin, and issue instructions through SCADA to feed back to the field monitoring station of the water conservancy full-element monitoring and control unit, so as to control the status of water conservancy facilities in the entire basin in real time and form a complete control closed loop. S9: After the rainfall ends, the results generated by the hydrological and hydrodynamic mechanism prediction model and the numerical prediction model are compared with the actual liquid level changes. The weight coefficients of each boundary condition in the numerical prediction model are optimized, and the relevant parameters in the hydrological and hydrodynamic mechanism prediction model are optimized to predict and warn of the risks of waterlogging, overflow and water quality in the basin.

[0007] Furthermore, the watershed network data in S1 is provided by the basic database of the watershed full-element monitoring and control unit, including information on sewage networks, sewage treatment plants, pumping stations, CSO storage tanks and enhanced treatment facilities, and waterways.

[0008] Furthermore, the monitoring data in S2 includes the influent flow and water quality of the sewage treatment plant, the liquid level in the forebay of the sewage treatment plant pumping station, the discharge flow and water quality of the sewage pumping station, the liquid level in front of the gate at the beginning of the intercepting culvert, the liquid level at the overflow gate at the end of the intercepting culvert, the influent or effluent flow and water quality of the CSO regulating tank, the liquid level at the end of the downstream pipeline network, and the discharge flow of the drainage pumping station.

[0009] Furthermore, the empirical data in S3 includes total rainfall, rainfall intensity, rainfall duration, initial liquid level in the pipeline network, opening and closing status of water conservancy facilities in the basin, and corresponding changes in pipeline and river or open channel liquid levels under the operational data. The operational data includes inflow and outflow flow rates and water quality data.

[0010] Furthermore, in S3, the formula for the rate of change of liquid level in the pipeline network at key points is: (1) (2) Among them, V i This represents the rate of increase in the liquid level in the pipeline at time i, in m / h. S represents the cross-sectional area of ​​the pipeline network, in meters (m²). 2 ; m represents the weighting coefficient affecting the change in liquid level in the pipeline network; R i This represents the rainfall intensity in the watershed at time i, expressed in mm / h. 'a' represents the weighting coefficient for rainfall intensity; Q represents the amount of wastewater treated by the wastewater treatment facility at time i, in meters. 3 / h; b represents the weighting coefficient for wastewater treatment volume; A0 represents the catchment area, in m². 2 ; H represents the height of the liquid level rise in the culvert, in meters (m). n represents the number of different rainfall intensities in the predicted rainfall event, expressed in units of [number]. t i The rainfall duration is given when the rainfall intensity is i in the watershed, in hours.

[0011] Furthermore, in S3, under different rainfall conditions, the system predicts the changes in the watershed network or culvert level based on rainfall intensity, initial water level in the pipeline network, and operating status of sewage treatment facilities, providing data reference for rationally scheduling the opening and closing of inter-basin connecting gates, storage and treatment facilities, and terminal overflow gates.

[0012] A watershed water environment prediction and early warning system, applicable to the watershed water environment prediction and early warning method, includes: The watershed full-element monitoring and control unit includes a professional database and a basic database. The professional database is used to receive real-time hydraulic and water quality monitoring data from sensors, rainfall forecasts from radar, and facility operation status from equipment. The basic database is used to collect basic geographic information, water conservancy facility information, and assessment and evaluation information from drawings and multimedia. The watershed prediction and early warning unit connects to and receives data sent by the watershed full-element monitoring and control unit. It is used to simulate the hydrology and water quality of sunny days and rainfall events in real time based on the monitoring data, including hydrological and hydrodynamic mechanism prediction models and numerical prediction models. Based on the prediction data of the two models, it provides early warning of the risks of possible waterlogging, overflow and water quality in the future, and publishes early warning information through the information sharing function of the platform. The basin joint regulation and control unit connects to and receives data sent by the basin prediction and early warning unit, and feeds back to the basin full-element monitoring and control unit. The basin joint regulation and control unit includes a real-time control and scheduling model and an empirical scheduling model, which are used to optimize the scheduling strategy and scheduling instructions based on the prediction results and early warning information, and to control the status of water conservancy facilities in the entire basin in real time.

[0013] Furthermore, the watershed-wide monitoring and control unit also includes: On-site monitoring stations are used to collect meteorological, hydrological, and water quality data in real time; On-site monitoring stations are used to collect and control the operating status of equipment and facilities in real time. The monitoring data acquisition and transmission layer is used to receive monitoring data sent by the field monitoring station and the field surveillance station, and send it to the professional database; The basic data layer is used to collect and organize experience data obtained through multiple channels and send it to the basic database.

[0014] A computer device includes a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the aforementioned prediction and early warning method.

[0015] The beneficial effects of this invention are reflected in: 1. This invention has developed a domestically produced dynamic simulation model for urban watershed water systems, formed a method for evaluating, predicting, and warning the integrated operation efficiency of plants, networks, and rivers based on a dual-model driven hydrological and hydrodynamic mechanism prediction model and a numerical prediction model, and developed a hydrological and water quality coordinated regulation technology for drainage systems based on a dual-model control of a real-time control and scheduling model and a numerical scheduling model. This has enabled joint scheduling and real-time control of multiple facilities including sources, networks, plants, stations, and rivers, meeting both the rapid response requirements of watershed scheduling and the complex operating conditions of the watershed. Thus, it has achieved intelligent management and control of the watershed water environment, and promoted the management and refined management of multiple objectives of the watershed, namely water security objectives, water environment objectives, and water ecological objectives.

[0016] 2. This invention is based on a dual-model driven approach using both numerical and mechanistic models. On one hand, the system platform receives real-time hydraulic and water quality monitoring data from sensors, rainfall forecasts from radar, and facility operation status monitoring data from equipment. It displays and analyzes the status of the entire basin in real time and evaluates key performance indicators. On the other hand, based on the monitoring data, it constructs hydrological and hydrodynamic mechanism prediction models and numerical prediction models to simulate the hydrology and water quality of sunny and rainy days in real time. Based on empirical data, it makes early predictions of potential future flooding, overflows, and water pollution, and issues early warning information. The system platform also uses optimization algorithms to optimize and generate optimized scheduling strategies and instructions based on the prediction results and early warning information, thereby controlling the status of water conservancy facilities throughout the basin in real time and forming a complete closed-loop computational logic. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the prediction and early warning method according to an embodiment of the present invention.

[0018] Figure 2 This is a structural block diagram of the prediction and early warning system according to an embodiment of the present invention.

[0019] Figure 3 This is a structural block diagram of a computer device according to an embodiment of the present invention.

[0020] The components in the attached diagram are labeled as follows: 1. Basin-wide monitoring and control unit; 2. Basin prediction and early warning unit; 3. Basin joint regulation and control unit; 4. Field monitoring station; 5. Field monitoring station; 6. Monitoring data acquisition and transmission layer; 7. Specialized database; 8. Basic data layer; 9. Basic database; 10. Hydrological and hydrodynamic mechanism prediction model; 11. Numerical prediction model; 12. Real-time control and scheduling model; 13. Empirical scheduling model. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. It should be noted that the meaning of "and / or" throughout the text includes three parallel solutions. Taking "A and / or B" as an example, it includes solution A, solution B, or a solution that simultaneously satisfies A and B.

[0022] See Figure 1 This invention provides a method for predicting and warning about the water environment in a watershed, comprising the following steps: S1: Based on existing watershed network data, combined with monitoring liquid levels and flow rates at key points of the network, and considering the watershed runoff patterns, and based on the operational logic between the watershed network, pumping stations, sewage treatment facilities, and rivers, the SWMM / MIKE / infoworks model is used to simulate the operational patterns of various drainage facilities in the watershed's "network-station-plant-river" system, and a watershed hydrological and hydrodynamic mechanism prediction model is constructed. S11: Based on the operational logic of various facilities in the "network-station-plant-river" within the watershed, and combined with the background conditions of the watershed, the catchment area is subdivided. The SWMM / MIKE / infoworks model is used to simulate the runoff generation and confluence process of surface runoff in the watershed catchment area, the hydraulic state of pipelines and open channels, and the diffusion of pollutants in open channels. In this way, a one-dimensional hydrological, hydrodynamic and water quality model of the watershed is established to evaluate the overflow of the upstream pipeline network and the restoration effect of river water quality. S12: After the hydrological, hydrodynamic and water quality model is established, the key parameters of the model are calibrated and the model is verified by selecting actual rainfall events. When the Nash coefficient is ≥0.6, the key parameter calibration requirements of the model are met. S2: Before or during rainfall, based on rainfall forecast data and real-time monitoring data provided by the professional database 7 of the watershed full-element monitoring and control unit 1, the established hydrological and hydrodynamic mechanism prediction model 10 is used to simulate and predict the liquid level changes in the watershed pipe network. S3: Using machine learning algorithms, analyze historical rainfall event data, establish a functional relationship between the operating status of different water facilities and the changes in pipe network liquid level at key control points in the basin under different rainfall conditions, and construct a numerical prediction model 11; S31: The formula for the rate of change of liquid level in the pipeline network at key points is: (1) (2) Among them, V i This represents the rate of increase in the liquid level in the pipeline at time i, in m / h. S represents the cross-sectional area of ​​the pipeline network, in meters (m²). 2 ; m represents the weighting coefficient affecting the change in liquid level in the pipeline network; R i This represents the rainfall intensity in the watershed at time i, expressed in mm / h. 'a' represents the weighting coefficient for rainfall intensity; Q represents the amount of wastewater treated by the wastewater treatment facility at time i, in meters. 3 / h; b represents the weighting coefficient for wastewater treatment volume; A0 represents the catchment area, in m². 2 ; H represents the height of the liquid level rise in the culvert, in meters (m). n represents the number of different rainfall intensities in the predicted rainfall event, expressed in units of [number]. t i The rainfall duration is given when the rainfall intensity is i in the watershed, in hours. S4: Before or during rainfall, the numerical prediction model 11 combines rainfall forecast data and real-time monitoring data, utilizes the established functional relationship, matches the corresponding weight coefficients, and predicts the changes in the watershed's key control points and pipeline network liquid levels. S5: Based on the hydrological and hydrodynamic mechanism prediction model 10 and the numerical prediction model 11, the platform predicts the time of overflow level occurrence. Considering the safety and immediate responsiveness of the scheduling, the prediction results of the two models are compared, and the prediction result of the model that first causes the overflow is selected as a reference for the scheduling strategy. S6: Based on the prediction and early warning results of the prediction model, the platform uses optimization algorithms to comprehensively consider water security and water environment objectives, based on the following scheduling rules, to generate scheduling instructions for various water facilities in the basin, forming the recommended scheduling scheme one; S61: Under sunny or light rain conditions: The watershed mainly relies on the full-load operation of the sewage treatment plant to control the liquid level in the underground culverts at a safe level, namely, the liquid level in the Huangxiao River underground culvert ≤18.5m and the liquid level in the Airport River underground culvert ≤18.3m. Under moderate rain conditions: (1) Huangxiao River Basin: ① When H = 18.1m and V i When the value is greater than 0, open the gate at the beginning of the Huangxiao River intercepting culvert and the Huangxiao River CSO regulating tank and enhanced treatment facilities; ②When H reaches 18.1m again and V i If the rainfall in the basin is less than 15 mm within the next 24 hours and the water level at the upstream waterlogging point is less than 0, the overflow gate at the end of the Huangxiao River intercepting culvert will be opened when H rises to 18.5 m. ③ When H < 18.0m and V i When the value is ≤0, close the gate at the beginning of the Huangxiao River intercepting sewage culvert; (2) Airport River Basin: ① When H = 18.0m and V i When the value is greater than 0, open the Changqing CSO regulating reservoir; ② When the Changqing CSO storage tank is full, H reaches 18.0m again and V i When the value is greater than 0, the airport river intercepting culvert and the airport river CSO regulating reservoir and enhanced treatment facilities will be put into operation. ③ When H≥18.3m or the rainfall in the basin within the next 24 hours is >15mm or the water level at the upstream waterlogging point h>0, open the overflow gate at the end of the intercepting sewage box culvert of the airport river; ④ When the rainfall ends and the rainfall in the basin is less than 15mm in the next 24 hours, close the overflow gate at the end of the intercepting sewage box culvert of the airport river; ⑤ After rainfall and runoff cease, when H < 18.0m and V i When the temperature is ≤0, close the gate at the beginning of the intercepting sewage culvert of the airport river; Under heavy rain or above conditions: (1) Huangxiao River Basin: After receiving the weather forecast, the sewage treatment capacity was increased and the liquid level in the culvert was lowered to below 17.8m. ① When H = 18.0m and V i When the value is greater than 0, open the gate at the beginning of the Huangxiao River intercepting culvert and the Huangxiao River CSO regulating tank and enhanced treatment facilities; ②When H rises again and reaches 18.0m, or h>0m, open the overflow gate at the end of the intercepting culvert to overflow to the end of the open channel; ③ If all the above facilities are activated but still cannot stop the rise in the liquid level of the culvert or alleviate the waterlogging situation, the steel dam gate at the end of the Huangxiao River culvert will be opened (dam collapse). ④ When H < 18.5m, V i When the rainfall in the basin is less than 25 mm within the next 24 hours, the steel dam gate (vertical dam) at the end of the Huangxiao River culvert should be closed first, followed by the overflow gate at the end of the Huangxiao River sewage interception box culvert. ⑤ After the rainfall ends and the runoff ends, when H < 18.0m and V i When the value is ≤0, close the gate at the beginning of the Huangxiao River intercepting sewage culvert.

[0023] (2) Airport River Basin: After receiving the weather forecast, the sewage treatment capacity was increased and the liquid level in the culvert was lowered to below 17.6m. ① When H = 18.0m and V i When the value is greater than 0, open the Changqing CSO regulating reservoir; ② When the Changqing CSO storage tank is full, H reaches 18.0m again and V i When the value is greater than 0, the airport river intercepting culvert and the airport river CSO regulating reservoir and enhanced treatment facilities will be put into operation. ③ When H rises again to 18.3m, or when the water level at the upstream waterlogging point h ≥ 0, open the overflow gate at the end of the intercepting culvert; ④ If all the above facilities are activated but still cannot stop the rise in the culvert level or alleviate the waterlogging situation, then open the steel dam gate at the end of the airport river culvert (dam collapse). ⑤ When the rainfall and runoff end, and the rainfall in the basin is less than 25 mm in the next 24 hours, the steel dam gate (vertical dam) at the end of the airport river culvert should be closed first, followed by the overflow gate at the end of the airport river intercepting sewage box culvert; ⑥ After the rainfall ends, when H < 17.9m and V iWhen the value is ≤0, close the gate at the beginning of the Huangxiao River intercepting sewage culvert.

[0024] S62: During rainfall, based on real-time monitoring data (including pipeline overflow flow and water quality data, and flow and water quality data of open channel outlets) at overflow gates or steel dam gates and open channel outlets, and based on real-time monitoring data at open channel monitoring points, zero-dimensional, one-dimensional or two-dimensional water quality models can be selected to predict changes in open channel water quality indicators according to the key points of river management. Taking the prediction of COD concentration at the Desheng Bridge section, upstream of the Huangxiao River open channel, as an example (one-dimensional model): The overflow flow rates at the overflow gate are Q1 and C1 (COD concentration), the overflow flow rates at the steel dam gate are Q2 and C2 (COD concentration), and the flow rates at the Huangxiao River open channel are Q3 and C3 (COD concentration). COD concentration:

[0025]

[0026] Where x represents the distance from the monitoring section to the beginning of the open channel, in meters; u represents the flow velocity of water in the open channel, in m / s; K represents the pollutant degradation rate, with units of 1 / d; Q represents flow rate, measured in meters (m). 3 / s; C represents COD concentration, with units of mg / L; S63: If the water quality forecast results show that the water quality of the open channel exceeds the standard, the drainage pumping station at the end of the open channel shall be activated to pump out water, or the ecological water replenishment of the open channel shall be increased.

[0027] S7: Based on rainfall forecasts and the operational status of water facilities, select and match historical rainfall event scheduling processes that match the background conditions of forecasted rainfall events from the historical rainfall event scheduling experience database, and generate scheduling scheme two; S8: During rainfall, management personnel, based on the actual rainfall situation and the operational status of water conservancy facilities in the basin, invoke dispatch instructions that are consistent with the actual operational status of the basin and issue them to each plant and station through the platform's SCADA system; S9: After the rainfall ends, the simulation results generated by the hydrological and hydrodynamic mechanism prediction model 10 and the numerical prediction model 11 are compared with the actual liquid level changes. The weight coefficients of each boundary condition in the numerical prediction model 11 are optimized, and the optimized weight coefficients that meet the accuracy requirements are included in the weight coefficient experience library of the numerical prediction model 11. The relevant parameters in the hydrological and hydrodynamic mechanism prediction model 10 are calibrated and verified, and the accuracy of the model prediction is continuously optimized to predict and warn of the risks of waterlogging, overflow and water quality in the basin.

[0028] This invention has developed a domestically produced dynamic simulation model for urban watershed water systems, and formed a method for evaluating, predicting, and warning the integrated operation efficiency of water treatment plants, networks, and rivers based on a dual-model driven approach of hydrological and hydrodynamic mechanism prediction model and numerical prediction model. It has also developed a hydrological and water quality coordinated regulation technology for drainage systems based on a dual-model control approach of real-time control and scheduling model and numerical scheduling model. This enables joint scheduling and real-time control of multiple facilities, including sources, networks, plants, stations, and rivers, which not only meets the rapid response requirements of watershed scheduling but also satisfies the complex operating conditions of the watershed. Thus, it realizes intelligent management and control of the watershed water environment and promotes the management and refined management of multiple objectives of the watershed, namely water security objectives, water environment objectives, and water ecology objectives.

[0029] See Figure 1 In this embodiment, the watershed network data in S1 is provided by the basic database 9 of the watershed full-element monitoring and control unit 1, including information about sewage network, sewage treatment plant, pumping station, CSO storage tank and enhanced treatment facilities, and river.

[0030] See Figure 1 In this embodiment, the monitoring data in S2 includes the influent flow and water quality of the sewage treatment plant, the liquid level in the forebay of the sewage treatment plant pumping station, the discharge flow and water quality of the sewage pumping station, the liquid level in front of the gate at the beginning of the intercepting culvert, the liquid level at the overflow gate at the end of the intercepting culvert, the influent or effluent flow and water quality of the CSO regulating tank, the liquid level at the end of the downstream pipeline network, and the discharge flow of the drainage pumping station.

[0031] See Figure 1 In this embodiment, the empirical data in S3 includes total rainfall, rainfall intensity, rainfall duration, initial liquid level of the pipeline network, opening and closing status of water conservancy facilities in the basin (gate at the beginning of the intercepting culvert, gate at the end of the intercepting culvert, and connecting gate within the basin) and the corresponding pipeline liquid level and river / open channel liquid level change data under the operational data. The operational data includes inflow and outflow flow and water quality data.

[0032] See Figure 1 In this embodiment, S3 further includes: S32: Based on the rainfall events, summarize the rainfall amount, the operation status of water conservancy facilities in the basin (hourly inflow of sewage treatment plants, opening and closing status of storage facilities and hourly inflow, and the corresponding rate of rise of the culvert level in the two rivers, etc., establish the correspondence between hourly rainfall intensity, inflow of sewage treatment facilities and the rate of rise of the culvert level in the downstream of the basin, and form an empirical database for the prediction and early warning module. S33: Based on historical rainfall event experience data, machine learning algorithms are used to establish a functional relationship between rainfall intensity, sewage treatment capacity (influent flow rate of sewage treatment plant, influent flow rate of storage facility, and connecting gate) and the rate of change of culvert level, forming a weighted coefficient database.

[0033] See Figure 2 This invention also provides a watershed water environment prediction and early warning system, applicable to the watershed water environment prediction and early warning method, comprising: The watershed full-element monitoring and control unit 1 includes a professional database 7 and a basic database 9. The professional database 7 is used to receive real-time hydraulic and water quality monitoring data from sensors, rainfall forecasts from radar, and facility operation status from equipment. The basic database 9 is used to collect basic geographic information, water conservancy facility information, and assessment and evaluation information from drawings and multimedia. The basin prediction and early warning unit 2 connects to and receives data sent by the basin full-element monitoring and control unit 1, and is used to simulate the hydrology and water quality of sunny days and rainfall events in real time based on the monitoring data. It includes a hydrological and hydrodynamic mechanism prediction model 10 and a numerical prediction model 11. Based on the prediction data of the two models, it provides early warning of the risks of possible waterlogging, overflow and water quality in the future, and publishes early warning information through the information sharing function of the platform. The basin joint regulation and control unit 3 connects to and receives data sent by the basin prediction and early warning unit 2, and feeds back to the basin full-element monitoring and control unit 1. The basin joint regulation and control unit 3 includes a real-time control and scheduling model 12 and an experience scheduling model 13, which are used to optimize the scheduling strategy and scheduling instructions by using optimization algorithms based on prediction results and early warning information, and to control the status of water conservancy facilities in the entire basin in real time.

[0034] This invention is driven by a dual-model approach based on numerical and mechanistic models. On one hand, the system platform receives real-time hydraulic and water quality monitoring data from sensors, rainfall forecasts from radar, and facility operation status monitoring data from equipment. It displays and analyzes the status of the entire basin in real time and evaluates key performance indicators. On the other hand, based on the monitoring data, it constructs hydrological and hydrodynamic mechanism prediction models and numerical prediction models to simulate the hydrology and water quality of sunny and rainy events in real time. Based on empirical data, it makes early predictions of possible future flooding, overflows, and water pollution, and issues early warning information. The system platform also uses optimization algorithms to optimize and generate optimized scheduling strategies and instructions based on the prediction results and early warning information, and controls the status of water conservancy facilities in the entire basin in real time, forming a complete closed-loop operation logic.

[0035] See Figure 2 In this embodiment, the watershed full-element monitoring and control unit 1 further includes: On-site monitoring station 4 is used to collect meteorological, hydrological and water quality data in real time; On-site monitoring station 5 is used to collect and control the operating status of equipment and facilities in real time; The monitoring data acquisition and transmission layer 6 is used to receive monitoring data sent by the on-site monitoring station 4 and the on-site monitoring station 5, and send it to the professional database 7; The basic data layer 8 is used to collect and organize experience data obtained through multiple channels and send it to the basic database 9.

[0036] In this embodiment, the monitoring data acquisition and transmission layer 6 collects monitoring data, such as water level, water quality, flow rate, and rainfall, from each of the field monitoring stations 4 via online monitoring communication. The data is then uploaded to the professional database 7 via remote communication methods, such as RTU and SCADA. The experience data obtained from the basic data layer 8 can be collected through data compilation, field surveys, etc. The system combines the professional database 7 and the basic database 9 for comprehensive analysis and prediction.

[0037] See Figure 3 This invention also provides a computer device, including a memory and a processor. The memory stores a computer program, which, when executed by the processor, causes the processor to perform the steps of the aforementioned prediction and early warning method. The steps of the prediction and early warning method here can be any of the steps in the prediction and early warning methods of the various embodiments described above.

[0038] In summary, this invention has developed a domestically produced dynamic simulation model for urban watershed water systems, formed a method for evaluating, predicting, and warning the integrated operation efficiency of water treatment plants, networks, and rivers based on a dual-model driven approach of hydrological and hydrodynamic mechanism prediction model and numerical prediction model, and developed a hydrological and water quality coordinated regulation technology for drainage systems based on a dual-model control approach of real-time control and scheduling model and numerical scheduling model. This has enabled joint scheduling and real-time control of multiple facilities including sources, networks, plants, stations, and rivers, meeting both the rapid response requirements of watershed scheduling and the complex operating conditions of the watershed. Thus, it has achieved intelligent management and control of the watershed water environment, promoting the management and refined management of multiple objectives of the watershed, namely water security objectives, water environment objectives, and water ecological objectives. This invention is driven by a dual-model approach based on numerical and mechanistic models. On one hand, the system platform receives real-time hydraulic and water quality monitoring data from sensors, rainfall forecasts from radar, and facility operation status monitoring data from equipment. It displays and analyzes the status of the entire basin in real time and evaluates key performance indicators. On the other hand, based on the monitoring data, it constructs hydrological and hydrodynamic mechanism prediction models and numerical prediction models to simulate the hydrology and water quality of sunny and rainy events in real time. Based on empirical data, it makes early predictions of possible future flooding, overflows, and water pollution, and issues early warning information. The system platform also uses optimization algorithms to optimize and generate optimized scheduling strategies and instructions based on the prediction results and early warning information, and controls the status of water conservancy facilities in the entire basin in real time, forming a complete closed-loop operation logic.

[0039] It should be understood that the examples and embodiments described herein are for illustrative purposes only and are not intended to limit the invention. Those skilled in the art can make various modifications or changes based on them. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the invention should be included within the protection scope of the invention.

Claims

1. A method for predicting and early warning of watershed water environment, characterized in that, Includes the following steps: S1: Based on the existing watershed network data, combined with the monitoring liquid level and flow rate at key points of the network, and combined with the watershed water catchment pattern, based on the operational logic between the watershed network, pumping stations, sewage treatment facilities and rivers, the SWMM / MIKE / infoworks model is used to simulate the operational pattern of various drainage facilities in the watershed "network-station-plant-river", and a watershed hydrological and hydrodynamic mechanism prediction model is constructed (10). S2: Before or during rainfall, based on rainfall forecast data and real-time monitoring data provided by the professional database (7) of the watershed full-element monitoring and control unit (1), the established hydrological and hydrodynamic mechanism prediction model (10) is used to simulate and predict the liquid level changes of the watershed pipe network. S3: Using machine learning algorithms, analyze historical rainfall event data, establish the functional relationship between the operating status of different water facilities under different rainfall conditions and the changes in the water level of the pipeline network at key control points in the basin, and construct a numerical prediction model (11). S4: Before or during rainfall, the numerical prediction model (11) combines rainfall forecast data and real-time monitoring data, and uses the established functional relationship to predict the changes in the watershed pipeline network liquid level; S5: Based on the hydrological and hydrodynamic mechanism prediction model (10) and the numerical prediction model (11), the platform predicts the time of overflow level occurrence. Considering the safety and immediate responsiveness of the scheduling, the prediction results of the two models are compared, and the prediction result of the model that first causes the overflow is selected as a reference for the scheduling strategy. S6: Based on the combined prediction results of the hydrological and hydrodynamic mechanism prediction model (10) and the numerical prediction model (11), and according to the scheduling rules formulated in the scheduling scheme, the optimization algorithm is used to find the best and construct the real-time control scheduling model (12). S7: Based on the scheduling experience database accumulated from previous rainfall events, pipeline liquid level, and plant facility operation status, an experience scheduling model is constructed (13). S8: Combining the real-time control and scheduling model (12) and the experience scheduling model (13), a recommended scheduling strategy and scheduling instructions are given. The management personnel determine the final scheduling plan based on the rainfall and the operating status of the water conservancy facilities in the basin, and issue instructions through SCADA to feed back to the field monitoring station (5) of the water conservancy full-element monitoring and control unit (1) in the basin, so as to control the status of the water conservancy facilities in the entire basin in real time and form a complete control closed loop. S9: After the rainfall ends, the results generated by the hydrological and hydrodynamic mechanism prediction model (10) and the numerical prediction model (11) are compared with the actual liquid level changes. The weight coefficients of each boundary condition in the numerical prediction model (11) are optimized, and the relevant parameters in the hydrological and hydrodynamic mechanism prediction model (10) are optimized to predict and warn of the risks of waterlogging, overflow and water quality in the basin.

2. The watershed water environment prediction and early warning method as described in claim 1, characterized in that, The watershed network data in S1 is provided by the basic database (9) of the watershed full-element monitoring and control unit (1), including information on sewage network, sewage treatment plant, pumping station, CSO storage tank and enhanced treatment facilities, and river.

3. The watershed water environment prediction and early warning method as described in claim 1, characterized in that, The monitoring data in S2 includes the influent flow and water quality of the sewage treatment plant, the liquid level in the forebay of the sewage treatment plant pumping station, the discharge flow and water quality of the sewage pumping station, the liquid level in front of the gate at the beginning of the intercepting culvert, the liquid level at the overflow gate at the end of the intercepting culvert, the influent or effluent flow and water quality of the CSO regulating tank, the liquid level at the end of the downstream pipeline network, and the discharge flow of the drainage pumping station.

4. The watershed water environment prediction and early warning method as described in claim 1, characterized in that, The empirical data in S3 includes total rainfall, rainfall intensity, rainfall duration, initial liquid level in the pipeline network, opening and closing status of watershed water facilities, and corresponding changes in pipeline and river or open channel liquid levels under the operational data. The operational data includes inflow and outflow flow rates and water quality data.

5. The watershed water environment prediction and early warning method as described in claim 1, characterized in that, In S3, the formula for the rate of change of liquid level in the key point pipeline network is: （1） （2） Among them, V i This represents the rate of increase in the liquid level in the pipeline at time i, in m / h. S represents the cross-sectional area of ​​the pipeline network, in meters (m²). 2 ; m represents the weighting coefficient affecting the change in liquid level in the pipeline network; R i This represents the rainfall intensity in the watershed at time i, expressed in mm / h. 'a' represents the weighting coefficient for rainfall intensity; Q represents the amount of wastewater treated by the wastewater treatment facility at time i, in meters. 3 / h; b represents the weighting coefficient for wastewater treatment volume; A0 represents the catchment area, in m². 2 ; H represents the height of the liquid level rise in the culvert, in meters (m). n represents the number of different rainfall intensities in the predicted rainfall event, expressed in units of [number]. t i The rainfall duration is given when the rainfall intensity is i in the watershed, in hours.

6. The watershed water environment prediction and early warning method as described in claim 1, characterized in that, In S3, under different rainfall conditions, the system predicts the changes in the watershed network or culvert level based on rainfall intensity, initial water level in the pipeline network, and operating status of sewage treatment facilities, providing data reference for the rational scheduling of the opening and closing of inter-basin connecting gates, storage and treatment facilities, and terminal overflow gates.

7. A watershed water environment prediction and early warning system, applicable to the watershed water environment prediction and early warning method as described in any one of claims 1-6, characterized in that, include: The basin-wide monitoring and control unit (1) includes a professional database (7) and a basic database (9). The professional database (7) is used to receive real-time hydraulic and water quality monitoring data from sensors, rainfall forecasts from radar, and facility operation status from equipment. The basic database (9) is used to collect basic geographic information, water conservancy facility information, and assessment and evaluation information from drawings and multimedia. The basin prediction and early warning unit (2) connects to and receives data sent by the basin full-element monitoring and control unit (1), and is used to simulate the hydrology and water quality of sunny days and rainfall events in real time based on the monitoring data, including the hydrological and hydrodynamic mechanism prediction model (10) and the numerical prediction model (11). Based on the prediction data of the two models, it provides early warning of the risks of waterlogging, overflow and water quality that may occur in the future, and publishes early warning information through the information sharing function of the platform. The basin joint regulation and control unit (3) connects to and receives data sent by the basin prediction and early warning unit (2), and feeds back to the basin full-element monitoring and control unit (1). The basin joint regulation and control unit (3) includes a real-time control and scheduling model (12) and an experience scheduling model (13), which are used to optimize the scheduling strategy and scheduling instructions by using optimization algorithms based on prediction results and early warning information, and to control the status of water conservancy facilities in the entire basin in real time.

8. The watershed water environment prediction and early warning system as described in claim 7, characterized in that, The basin-wide monitoring and control unit (1) also includes: On-site monitoring stations (4) are used to collect meteorological, hydrological and water quality data in real time; On-site monitoring station (5) is used to collect and control the operating status of equipment and facilities in real time; The monitoring data acquisition and transmission layer (6) is used to receive monitoring data sent by the field monitoring station (4) and the field monitoring station (5) and send it to the professional database (7). The basic data layer (8) is used to collect and organize experience data obtained through multiple channels and send it to the basic database (9).

9. A computer device comprising a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the method as claimed in any one of claims 1-6.

Images