Rainwater and accident pool combined system based on digital twinning

Abstract

The application is based on a rainwater and emergency pool combined system based on digital twinning, relates to the cross technical field of environmental engineering and intelligent control, includes a physical system, the physical system includes a combined pool, the combined pool is communicated with a rainwater pipe network and an emergency wastewater collection pipe network, a sensor group and an actuator group are uniformly arranged on the rainwater pipe network and the emergency wastewater collection pipe network; a digital system, the digital system is in communication connection with the physical system through a data interface; firstly, through the prospective dynamic capacity allocation of digital twinning and model predictive control, the application optimizes the utilization of limited pool capacity in the time dimension, changes the two functions of rainwater regulation and emergency from static segmentation to dynamic coordination, and improves the comprehensive utilization efficiency and emergency preparation level of the facility; secondly, based on real-time deduction and a rapid response mechanism of critical conditions, the optimal disposal strategy can be automatically generated and executed at the moment of emergency alarm, realizing the mode change from passive reception to active defense.

Description

Technical Field

[0001] This application relates to the interdisciplinary field of environmental engineering and intelligent control, and in particular to a combined rainwater and emergency storage system based on digital twins. Background Technology

[0002] In the current field of water treatment and industrial environmental risk prevention and control, in order to take into account both rainwater management and emergency storage of wastewater, the combined construction of rainwater storage tanks and emergency tanks has become a common engineering model that saves land and intensively invests. However, traditional combined tanks have obvious limitations in design and operation. The core problem lies in the static separation of functions and the passive lag in response. Specifically, traditional combined pools typically use physical partitions or fixed weirs to permanently divide the pool volume into two parts: one part for storing emergency wastewater and the other for regulating rainwater. This static zoning model means that the capacity between the two core functions cannot be dynamically adjusted according to actual needs. During the rainy season when there are no accidents, the valuable emergency pool volume is idle for a long time, resulting in a waste of resources. In extreme conditions where a sudden accident occurs and rainfall occurs, the fixed volume allocated to rainwater regulation may be squeezed out, resulting in a serious shortage of emergency capacity and a huge risk of pollution leakage. Furthermore, the operation of existing systems relies heavily on preset thresholds and human experience, lacking the ability to proactively predict and collaboratively optimize rainfall processes and accident risks. The response chain from accident occurrence and human judgment to manual operation of gates and pumps is too long, making it difficult to meet the modern emergency requirements of minute-level rapid response. Although technologies such as digital twins and the Internet of Things have been applied in drainage systems in recent years, they are mostly limited to data monitoring and status display, failing to form a closed-loop control that is deeply coupled with physical control equipment (such as adjustable weirs and pumps) and optimized in real time, making it difficult to achieve proactive and collaborative operation based on prediction. Summary of the Invention

[0003] To address the aforementioned issues, this application provides a combined rainwater and emergency pool system based on digital twins.

[0004] The rainwater and emergency storage pool integrated system based on digital twins provided in this application adopts the following technical solution: The integrated rainwater and emergency storage system based on digital twins includes: The physical system includes a combined pool, which is connected to a rainwater pipe network and an emergency wastewater collection pipe network. Sensor groups and actuator groups are evenly distributed on the rainwater pipe network and the emergency wastewater collection pipe network. A digital system, which communicates with the physical system via a data interface, comprising: The data fusion module is configured to acquire and fuse monitoring data from the sensor group, external weather forecast data, and factory production risk data in real time to form a system state dataset with a unified spatiotemporal reference. A coupled prediction model is communicatively connected to the data fusion module. The coupled prediction model is a hydraulic and water quality coupled prediction model, which is configured to: calculate in real time the dynamic capacity allocation boundary of the combined pool and the critical conditions for switching between rainwater storage mode and emergency response mode. A collaborative control module is communicatively connected to the coupled prediction model and the actuator group. The collaborative control module is configured to generate collaborative scheduling instructions based on the dynamic capacity allocation boundary and the critical conditions, and control the flow direction, flow rate and partitioning of the liquid in the combined pool through the actuator group.

[0005] As a preferred technical solution of this application, the combined pool is divided into two functionally dynamically interchangeable zones by adjustable weirs or valves. The sensor group includes water quality sensors and flow meters installed at the inlet of the rainwater pipe network, the inlet of the emergency wastewater collection pipe network, and in the two zones. The two zones are connected by a connecting pipe. The actuator group includes an electric actuator, a first water pump, and a second water pump. The electric actuator is used to control the adjustable weirs and valves. The first water pump is installed on the connecting pipe and is used to actively drive the water body to transfer between the two zones under the command of the collaborative control module. The second water pump is installed on the rainwater pipe network and the emergency wastewater collection pipe network and is used to regulate or cut off the water flow entering the combined pool under the command of the collaborative control module.

[0006] As a preferred technical solution of this application, the coupled prediction model includes a mutually coupled rainwater runoff sub-model, a pollutant migration and transformation sub-model, and a pool hydraulic sub-model. The dynamic capacity allocation boundary is a capacity allocation curve that changes over time based on the predicted rainfall process line and the real-time plant accident risk level.

[0007] As a preferred technical solution of this application, the collaborative control module is configured as follows: When the predicted initial rainwater pollution load exceeds the first threshold and the dynamic reserve capacity for emergency response calculated by the coupled prediction model is lower than the second threshold, a first type of control instruction is generated to guide some of the polluted rainwater to a designated zone for pretreatment. When a factory accident alarm signal is received, the quantity and quality of the accident wastewater are quickly estimated based on the coupled prediction model, and a second type of control command is generated accordingly. The second type of control command includes shutting down or adjusting the rainwater inlet facility and starting a dedicated collection and treatment process for the accident wastewater.

[0008] As a preferred technical solution of this application, the output end of the collaborative control module is connected to an interactive interface, which is configured to display the system status dataset, the dynamic capacity allocation boundary, the critical conditions, and the collaborative scheduling instructions in real time.

[0009] As a preferred technical solution of this application, an operation method for a combined rainwater and emergency storage system based on digital twins includes the following steps: S1 acquires and merges multi-source data in real time to form a unified dataset that reflects the overall state of the system; S2, input the unified dataset into the preset hydraulic and water quality coupling prediction model to calculate the dynamic capacity allocation boundary and critical conditions for mode switching of the combined pool in the current and future period. S3. Based on the dynamic capacity allocation boundary and the critical conditions, assess the capacity competition risk and the timing of the operating condition switching, and generate corresponding collaborative scheduling instructions. S4, the coordinated scheduling instruction is sent to the execution mechanism of the physical system to control the actions of each zone, valve and water pump in the combined pool, so as to realize the dynamic coordinated operation of rainwater storage and accident handling.

[0010] As a preferred technical solution of this application, in S2, the calculation of the dynamic capacity allocation boundary specifically involves: simulating the rainwater inflow and water quality process lines under future rainfall events based on meteorological forecast data; assessing the accident risk level and potential accident wastewater characteristics in future periods based on plant production plans and material storage data; and using the overall safe operation of the combined pool as a constraint, performing dynamic capacity allocation optimization between the rainwater inflow process line and the potential accident wastewater characteristics to generate the dynamic capacity allocation boundary curve.

[0011] As a preferred technical solution of this application, in S3, more specifically: during non-accident periods, according to the dynamic capacity allocation boundary curve, the actuator is controlled in advance to empty a portion of the volume of the accident pool as a reserve space for rainwater storage. Upon receiving an accident alarm, the coupled prediction model is immediately invoked to simulate the diffusion and treatment process of the accident wastewater in the combined pool, using the current state as the initial condition. The coordinated scheduling command is dynamically adjusted according to the simulation results to maximize the overall impact resistance of the system while treating the accident wastewater.

[0012] As a preferred technical solution of this application, after S4, the digital system records historical operating data, model prediction results and actual execution effects. Based on the comparison results, the system uses machine learning algorithms to adaptively calibrate and optimize the parameters of the coupled prediction model or the thresholds configured in the collaborative control module.

[0013] In summary, this application includes the following beneficial technical effects: This application firstly optimizes the utilization of limited pool capacity in the time dimension through forward-looking dynamic capacity allocation using digital twins and model predictive control, transforming the two major functions of rainwater storage and emergency response from static separation to dynamic synergy, thereby improving the overall utilization efficiency and emergency preparedness level of the facilities. Secondly, based on a rapid response mechanism for real-time simulation and critical conditions, the optimal handling strategy can be automatically generated and executed the instant an accident alarm is triggered, realizing a shift from passive reception to proactive defense, greatly shortening the emergency response time and effectively curbing the spread of pollution. Thirdly, through the setting of hardware interlocks and a three-level degradation strategy, it ensures that in any fault situation, from digital system failure to plant-wide power outage, the system can unconditionally prioritize the safe containment of accident wastewater through preset mechanical and logical redundancy, avoiding the risk of hazardous material leakage. Finally, by optimizing scheduling to reduce equipment start-up and shutdown frequency and energy consumption, and by continuously improving control accuracy based on machine learning self-calibration capabilities, the application achieves reduced operation and maintenance costs and intelligent decision-making throughout the entire life cycle. Attached Figure Description

[0014] Figure 1 This is the system architecture diagram of the combined rainwater and emergency storage tank in this application; Figure 2 This application Figure 1 Mid-system adaptive calibration and optimization architecture diagram; Figure 3 This is a flowchart of the operation method of the combined rainwater and emergency storage system of this application; Figure 4 This is a flowchart of the three-level downgrade strategy method for this application. Detailed Implementation

[0015] The following is in conjunction with the appendix Figure 1-4 This application will be described in further detail.

[0016] See Figure 1-4 A digital twin-based integrated rainwater and emergency storage system includes: The physical system includes a combined pool, which is connected to a rainwater pipe network and an emergency wastewater collection pipe network. Sensor groups and actuator groups are installed on both the rainwater pipe network and the emergency wastewater collection pipe network.

[0017] The combined pool is divided into two dynamically interchangeable functional zones by adjustable weirs or valves. The sensor group includes water quality sensors and flow meters installed at the inlet of the rainwater pipe network, the inlet of the emergency wastewater collection pipe network, and in the two zones. The two zones are connected by connecting pipes. The actuator group includes an electric actuator, a first water pump, and a second water pump. The electric actuator is used to control the adjustable weirs and valves. The first water pump is installed on the connecting pipe and is used to actively drive the water body to transfer between the two zones under the command of the collaborative control module. The second water pump is installed on the rainwater pipe network and the emergency wastewater collection pipe network and is used to regulate or cut off the water flow into the combined pool under the command of the collaborative control module.

[0018] The combined pool is divided into at least two main functional areas (Area A and Area B) by one or more liftable weir gates or openable and closed valves. The functions of these two areas (rainwater regulation or emergency wastewater storage) are not fixed and can be dynamically changed according to instructions. Zone A and Zone B are connected by a connecting pipeline. A first water pump is installed on the connecting pipeline. The first water pump is used to actively and quantitatively transfer water between the two zones under control commands to achieve rapid capacity allocation. Online flow meters and water quality sensors (such as pH, conductivity, COD, and concentration of characteristic pollutants) are installed at the inlet of the rainwater pipe network and the inlet of the emergency wastewater collection pipe network, respectively. Liquid level gauges and water quality sensors are installed inside Zone A and Zone B, respectively. The first water pump is used for water transfer between sections. The second water pump is installed in the rainwater pipe network and the emergency wastewater collection pipe network to regulate or cut off the water flow into the combined pool. The electric actuator is used to control the raising and lowering of the adjustable weir gate or the opening and closing of the valve. The critical weirs or valves in this application employ fail-safe actuators (such as automatically resetting to a safe position in the event of power failure / signal loss) and are equipped with hardware interlocks to ensure priority protection for the containment of accidental wastewater in extreme circumstances.

[0019] Digital systems, which communicate with physical systems through data interfaces, include: The data fusion module is configured to acquire and fuse monitoring data from the sensor group, external weather forecast data, and plant production risk data in real time to form a system status dataset with a unified spatiotemporal reference.

[0020] The data fusion module accesses and integrates real-time monitoring data from the sensor group of the physical system, refined forecast data (future rainfall process line) from the meteorological department, and production risk data (such as production plan, inventory of high-risk materials, and equipment status) from the plant information system. The data fusion module performs time synchronization, spatial matching, and formatting processing on the multi-source data and outputs a unified system status dataset.

[0021] The coupled prediction model is connected to the data fusion module. It is a hydraulic and water quality coupled prediction model, which is configured to: calculate the dynamic capacity allocation boundary of the combined reservoir in real time and the critical conditions for switching between rainwater storage mode and accident response mode. The coupled prediction model includes mutually coupled rainwater runoff sub-model, pollutant migration and transformation sub-model and reservoir hydraulic sub-model. The dynamic capacity allocation boundary is a capacity allocation curve that changes over time based on the predicted rainfall process line and the real-time plant accident risk level.

[0022] The coupled prediction model is a mechanism-data hybrid model that integrates the stormwater runoff sub-model, the pollutant migration and transformation sub-model, and the pool hydraulic sub-model. The core function of the coupled prediction model is to perform the following two types of key calculations based on the system state dataset and future prediction data: Calculate the dynamic capacity allocation boundary: Based on the future rainfall process line and the plant area accident risk prediction, with the total pool capacity and safety regulations as constraints, an optimization algorithm is used to calculate a curve that changes over time, indicating how much capacity should be reserved at each future moment to deal with potential accidents (V_reserve(t)) and how much is left to regulate rainwater (V_storm(t) = V_total - V_reserve(t)). Simulating critical conditions and accident scenarios: The real-time calculation system determines the critical conditions for switching between rainwater storage mode and accident response mode. Once an accident alarm is received, the coupled prediction model can quickly simulate the evolution of the wastewater volume and quality from the current pool state, providing a basis for emergency control decisions.

[0023] The collaborative control module communicates with the coupled prediction model and the actuator group. The collaborative control module is configured to generate collaborative scheduling instructions based on dynamic capacity allocation boundaries and critical conditions, and control the flow direction, flow rate, and zoning of the liquid within the combined pool through the actuator group. The collaborative control module is configured as follows: When the predicted initial rainwater pollution load exceeds the first threshold and the dynamic reserve capacity for emergency response calculated by the coupled prediction model is lower than the second threshold, a first type of control instruction is generated to guide some of the polluted rainwater to a designated zone for pretreatment. When a factory accident alarm signal is received, the quantity and quality of the accident wastewater are quickly estimated based on the coupled prediction model, and a second type of control command is generated accordingly. The second type of control command includes shutting down or adjusting the rainwater inlet facility and starting a dedicated collection and treatment process for the accident wastewater.

[0024] The collaborative control module receives the dynamic capacity allocation boundary and critical conditions output by the coupled prediction model, combines them with real-time data, generates specific and executable collaborative scheduling instructions, and sends them to the execution mechanism group of the entity system. The control logic of the collaborative control module includes preventive scheduling, emergency diversion of polluted rainwater, and rapid response to accidents.

[0025] The output of the collaborative control module has an interactive interface, which is configured to display real-time system status datasets, dynamic capacity allocation boundaries, critical conditions, and collaborative scheduling instructions.

[0026] The human-machine interface displays the system's overall status in real time in a graphical manner, including dynamic capacity allocation boundary curves, early warning information, control commands, and execution effects, providing operators with access to monitoring, querying, and manual intervention.

[0027] An operation method for a combined rainwater and emergency storage system based on digital twins includes the following steps: S1 acquires and merges multi-source data in real time to form a unified dataset that reflects the overall state of the system.

[0028] The system runs continuously, with the data fusion module collecting, cleaning, and fusing data from sensors, meteorological servers, and plant databases in real time to generate a system status dataset with a unified spatiotemporal benchmark, thus constructing a real-time mirror of the physical entity in the digital space.

[0029] S2. Input the unified dataset into the pre-set hydraulic and water quality coupled prediction model to calculate the dynamic capacity allocation boundary and critical conditions for mode switching of the combined pool in the current and future periods. The calculation of the dynamic capacity allocation boundary is as follows: based on meteorological forecast data, simulate the rainwater inflow and water quality process lines under future rainfall events; based on the plant production plan and material storage data, assess the accident risk level and potential accident wastewater characteristics in the future period; with the total safe operation of the combined pool as a constraint, perform dynamic capacity allocation optimization between the rainwater inflow process line and the potential accident wastewater characteristics to generate the dynamic capacity allocation boundary curve.

[0030] This step uses the Model Predictive Control (MPC) framework to solve a rolling time-domain optimization problem to generate dynamic capacity allocation boundaries. The specific process is as follows: Starting from the current time t0, optimize for the next T hours (e.g., T=24); Set the volume V_reserve(t) that should be reserved for emergency response in each discrete time period (e.g., every 15 minutes) as the main decision variable; Minimize the risk gap of insufficient reserved capacity when the accident risk is high. The formula is: minimize Σ_tmax(0, R(t)-β*V_reserve(t)), where R(t) is the quantified risk value at time t and β is the conversion coefficient.

[0031] To maximize the utilization value of rainwater storage capacity, we need to maximize Σ_t(V_storm(t)*I_rain(t)), where I_rain(t) is 1 when rainfall is predicted, and 0 otherwise.

[0032] Minimize the frequency of equipment operation and energy consumption caused by volume allocation adjustment, i.e. minimize Σ_t|V_reserve(t)-V_reserve(t-1)|.

[0033] The constraints of the above process include: Volume constraint: V_min_accident≤V_reserve(t)≤V_total; Hydraulic constraint: V_storm(t) must be able to safely contain the inflow of rainwater during time period t as predicted by the stormwater runoff sub-model; Equipment capacity constraint: The adjustment rate of V_reserve(t) must not exceed the actual capacity of equipment such as water pumps.

[0034] Finally, in each decision cycle (e.g., every 5 minutes), the coupled prediction model uses the latest system state dataset and external prediction data to call an optimization solver (e.g., interior point method, genetic algorithm) to solve the above model, obtaining the optimal V_reserve*(t) sequence for the next T hours. The first value of this sequence is taken as the capacity target that should be executed immediately, and the complete V_reserve*(t) curve is output as the dynamic capacity allocation boundary to guide subsequent control.

[0035] S3 assesses capacity competition risks and operating condition switching opportunities based on dynamic capacity allocation boundaries and critical conditions, and generates corresponding collaborative scheduling instructions. During non-accident periods, based on the dynamic capacity allocation boundary curve, it controls the actuators to empty a portion of the accident pool volume in advance as a reserve space for rainwater storage. Upon receiving an accident alarm, it immediately invokes the coupled prediction model, using the current state as the initial condition, to simulate the diffusion and treatment process of accident wastewater in the combined pool, and dynamically adjusts the collaborative scheduling instructions based on the simulation results, so as to maximize the overall impact resistance of the system while treating accident wastewater.

[0036] Based on the dynamic boundary curve, the collaborative control module controls the first water pump, the liftable weir gate, or the openable valve in advance to adjust the water distribution in areas A and B, reserving optimal space for high-probability events (such as upcoming rainfall or high-risk operations). When the initial rainwater pollution load is detected to exceed the set threshold, and the coupled prediction model calculates that the current emergency reserve capacity is lower than the safety threshold, an instruction is generated to guide the highly polluted rainwater to a designated zone for isolation and pretreatment to protect the core emergency capacity. Upon receiving an accident alarm, a rapid simulation is immediately triggered, and instructions are generated based on the results: close or reduce the rainwater inlet, fully open the accident wastewater inlet, and divert the accident wastewater to a pre-prepared zone to initiate a dedicated treatment process.

[0037] S4 sends the coordinated scheduling instructions to the execution mechanism of the physical system to control the actions of each zone, valve and water pump in the combined pool, so as to realize the dynamic coordinated operation of rainwater storage and accident handling.

[0038] The actuator group precisely executes commands, adjusts the status of weirs, valves or pumps, changes the water flow path and zone liquid level, realizes dynamic capacity allocation and smooth switching of operating conditions. The execution results are captured by sensors in real time and fed back to S1 to start the next control cycle.

[0039] The digital system records historical operational data, model prediction results, and actual execution effects. Based on the comparison results, machine learning algorithms are used to adaptively calibrate and optimize the parameters of the coupled prediction model or the thresholds configured in the collaborative control module.

[0040] During continuous operation, the system automatically records all historical data, including model predictions, control commands, and actual results. It periodically uses machine learning algorithms (such as time series analysis and Bayesian updates) to compare the deviations between predictions and actual results. It automatically calibrates and optimizes the parameters of the coupled prediction model (such as runoff coefficient and attenuation coefficient) and the decision thresholds of the collaborative control module, so that the digital twin continues to evolve and becomes closer to the real characteristics of the physical entity.

[0041] To ensure the basic safety of the system under any failure, a three-level degradation strategy is designed: Level 1 Degradation (Digital System Core Failure): When the optimization model or main control program fails, the local logic controller (PLC) takes over control based on preset fixed threshold rules (such as "if the emergency wastewater flow rate is greater than the threshold, then the rainwater inlet is cut off"). Level 2 Degradation (Control Network Interruption): When control commands cannot be issued, critical equipment (such as emergency discharge pumps to prevent overflow) is directly controlled by local sensors (such as automatic start when the liquid level is too high), and core valves or weirs rely on hardware interlocks to maintain a safe state (such as emergency wastewater valves and rainwater valves not being fully open at the same time). Level 3 Degradation (Power Loss to the Entire System): When all power sources fail, the fail-safe actuators of critical valves or weirs automatically reset them to preset safe positions (e.g., if the emergency wastewater inlet valve fails and is fully open, the rainwater inlet valve fails and is closed, or the connecting weir fails and is open), ensuring that the combined pool, as a whole, can preferentially accommodate emergency wastewater by gravity flow and prevent the leakage of hazardous substances.

[0042] This application first optimizes the utilization of limited pool capacity in the time dimension through forward-looking dynamic capacity allocation using digital twins and model predictive control, transforming the two major functions of rainwater storage and emergency response from static separation to dynamic synergy, thereby improving the overall utilization efficiency and emergency preparedness level of the facility. Secondly, the rapid response mechanism based on real-time simulation and critical conditions can automatically generate and execute the optimal handling strategy the moment an accident alarm is triggered, realizing the transformation from passive reception to active defense, greatly shortening the emergency response time and effectively curbing the spread of pollution. Then, through the set hardware interlocks and three-level degradation strategy, it is ensured that in any fault situation from digital system failure to plant-wide power outage, the system can unconditionally prioritize the safe containment of accident wastewater through preset mechanical and logical redundancy, avoiding the risk of hazardous material leakage. Finally, by optimizing scheduling to reduce the frequency of equipment start-ups and shutdowns and energy consumption, and by continuously improving control accuracy based on machine learning self-calibration capabilities, the operation and maintenance costs can be reduced and decision-making can be made more intelligent throughout the entire life cycle.

[0043] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A combined rainwater and emergency storage system based on digital twins, characterized in that, include: The physical system includes a combined pool, which is connected to a rainwater pipe network and an emergency wastewater collection pipe network. Sensor groups and actuator groups are evenly distributed on the rainwater pipe network and the emergency wastewater collection pipe network. A digital system, which communicates with the physical system via a data interface, comprising: The data fusion module is configured to acquire and fuse monitoring data from the sensor group, external weather forecast data, and factory production risk data in real time to form a system state dataset with a unified spatiotemporal reference. A coupled prediction model is communicatively connected to the data fusion module. The coupled prediction model is a hydraulic and water quality coupled prediction model, which is configured to: calculate in real time the dynamic capacity allocation boundary of the combined pool and the critical conditions for switching between rainwater storage mode and emergency response mode. A collaborative control module is communicatively connected to the coupled prediction model and the actuator group. The collaborative control module is configured to generate collaborative scheduling instructions based on the dynamic capacity allocation boundary and the critical conditions, and control the flow direction, flow rate and partitioning of the liquid in the combined pool through the actuator group.

2. The rainwater and emergency storage pool integrated system based on digital twins according to claim 1, characterized in that, The combined pool is divided into two dynamically interchangeable functional zones by adjustable weirs or valves. The sensor group includes water quality sensors and flow meters installed at the inlet of the rainwater pipe network, the inlet of the emergency wastewater collection pipe network, and in the two zones. The two zones are connected by a connecting pipe. The actuator group includes an electric actuator, a first water pump, and a second water pump. The electric actuator is used to control the adjustable weirs and valves. The first water pump is installed on the connecting pipe and is used to actively drive the water body to transfer between the two zones under the command of the collaborative control module. The second water pump is installed on the rainwater pipe network and the emergency wastewater collection pipe network and is used to regulate or cut off the water flow entering the combined pool under the command of the collaborative control module.

3. The rainwater and emergency storage pool integrated system based on digital twins according to claim 1, characterized in that, The coupled prediction model includes a rainwater runoff sub-model, a pollutant migration and transformation sub-model, and a pool hydraulic sub-model that are coupled with each other. The dynamic capacity allocation boundary is a capacity allocation curve that changes over time, calculated based on the predicted rainfall process line and the real-time plant accident risk level.

4. The rainwater and emergency storage pool integrated system based on digital twins according to claim 3, characterized in that, The collaborative control module is configured as follows: When the predicted initial rainwater pollution load exceeds the first threshold and the dynamic reserve capacity for emergency response calculated by the coupled prediction model is lower than the second threshold, a first type of control instruction is generated to guide some of the polluted rainwater to a designated zone for pretreatment. When a factory accident alarm signal is received, the quantity and quality of the accident wastewater are quickly estimated based on the coupled prediction model, and a second type of control command is generated accordingly. The second type of control command includes shutting down or adjusting the rainwater inlet facility and starting a dedicated collection and treatment process for the accident wastewater.

5. The rainwater and emergency storage pool integrated system based on digital twins according to claim 1, characterized in that, The output of the collaborative control module is connected to an interactive interface, which is configured to display the system status dataset, the dynamic capacity allocation boundary, the critical conditions, and the collaborative scheduling instructions in real time.

6. An operation method for a combined rainwater and emergency storage system based on digital twins, the combined rainwater and emergency storage system based on digital twins according to any one of claims 1-6, characterized in that, Includes the following steps: S1 acquires and merges multi-source data in real time to form a unified dataset that reflects the overall state of the system; S2, input the unified dataset into the preset hydraulic and water quality coupling prediction model to calculate the dynamic capacity allocation boundary and critical conditions for mode switching of the combined pool in the current and future period. S3. Based on the dynamic capacity allocation boundary and the critical conditions, assess the capacity competition risk and the timing of the operating condition switching, and generate corresponding collaborative scheduling instructions. S4, the coordinated scheduling instruction is sent to the execution mechanism of the physical system to control the actions of each zone, valve and water pump in the combined pool, so as to realize the dynamic coordinated operation of rainwater storage and accident handling.

7. The operation method of a combined rainwater and emergency storage system based on digital twins according to claim 6, characterized in that, In S2, the calculation of the dynamic capacity allocation boundary specifically involves: simulating the rainwater inflow and water quality process lines under future rainfall events based on meteorological forecast data; assessing the accident risk level and potential accident wastewater characteristics in future periods based on plant production plans and material storage data; and using the overall safe operation of the combined pool as a constraint, performing dynamic capacity allocation optimization between the rainwater inflow process line and the potential accident wastewater characteristics to generate the dynamic capacity allocation boundary curve.

8. The operation method of a combined rainwater and emergency storage system based on digital twins according to claim 7, characterized in that, More specifically in S3: During non-accident periods, based on the dynamic capacity allocation boundary curve, the actuator is controlled in advance to empty a portion of the accident pool volume as a reserve space for rainwater storage. Upon receiving an accident alarm, the coupled prediction model is immediately invoked to simulate the diffusion and treatment process of accident wastewater in the combined pool using the current state as the initial condition. The coordinated scheduling command is dynamically adjusted based on the simulation results to maximize the overall impact resistance of the system while treating accident wastewater.

9. The operation method of a combined rainwater and emergency storage system based on digital twins according to claim 6, characterized in that, After S4, the digital system records historical operating data, model prediction results, and actual execution effects. Based on the comparison results, it uses machine learning algorithms to adaptively calibrate and optimize the parameters of the coupled prediction model or the thresholds configured in the collaborative control module.

Images