Energy-saving control system and method for water supply-water distribution cooperative scheduling of water plant
By constructing a mixed water supply prediction model and a tidal coordinated water intake allocation model, the coordinated scheduling of water supply and distribution is realized, which solves the problems of low prediction accuracy and high energy consumption of traditional water supply systems when facing fluctuations in water demand and emergencies, and achieves efficient water supply scheduling and energy-saving effects.
Patent Information
- Application Number
- CN202511662067.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-13
- Publication Date
- 2026-02-17
AI Technical Summary
Traditional water supply systems have low prediction accuracy when facing nonlinear fluctuations in water demand and sudden events. Independent optimization of the water supply and distribution ends leads to high energy consumption and fails to effectively utilize the energy-saving benefits brought by tidal levels.
A mixed water supply prediction model was constructed, which was combined with a tidal coordinated water intake allocation model. The water intake flow was scheduled in real time through a multi-objective optimization model. The model integrates tidal level, water supply prediction and clear water tank level constraints to achieve coordinated scheduling of water supply and distribution.
It improves the accuracy of water supply forecasting, maximizes the use of tidal potential energy, reduces the power consumption of water intake pumping stations, and ensures water supply safety and operational efficiency.
Smart Images
Figure CN121543943A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of smart water technology, specifically to an energy-saving control system and method for coordinated scheduling of water supply and distribution in water plants. Background Technology
[0002] The refined scheduling of urban water supply systems faces three major challenges: long-term nonlinear trends, multi-period coupling, and frequent emergencies. Traditional water supply systems suffer from the following problems: 1. Water demand fluctuates nonlinearly due to factors such as weather and seasons, and is also accompanied by sudden water supply scheduling needs such as pipe bursts and fires. It is difficult to achieve accurate hourly forecasts based solely on historical data. Currently, water plant water supply forecasts mostly rely on single time series models such as ARIMA and LSTM, which cannot simultaneously handle multiple disturbances such as sudden demand during holidays, pressure scheduling instructions, and sudden weather changes, resulting in low forecast accuracy.
[0003] 2. The water supply end (water delivery pumping station) and the water distribution end (raw water pumping station) are usually optimized independently. The water plant uses the water level in the clear water tank as the control target, while the pipeline network uses the terminal pressure as the benchmark, and does not consider the energy-saving benefits of tidal liquid level to the water intake pumping station. Summary of the Invention
[0004] The purpose of this application is to provide an energy-saving control system and method for coordinated scheduling of water supply and distribution in water plants. A real-time rolling optimization model that integrates tidal level, water supply prediction, and clear water tank level constraints aims to improve prediction accuracy and maximize the utilization of tidal potential energy for energy saving and consumption reduction.
[0005] To achieve the above objectives, the technical solution adopted in this application is as follows: Energy-saving control methods for coordinated scheduling of water supply and distribution in water plants include: S1: Construct a mixed water supply prediction model, including: The predictive model, based on historical flow and pressure data at the water supply end, weather data, real-time dispatch instructions, and emergencies, generates future water supply forecasts by combining predictive models. The instantaneous water volume calculation model combines pressure commands and real-time data to calculate the target instantaneous water supply demand; S2: Obtain the dynamic liquid level safety range and absolute safety red line of the clear water tank; S3: Construct a tidal coordinated water intake allocation model. Using the instantaneous and future demand of S1, the liquid level constraint of S2, the real-time tidal liquid level, and the clear water tank liquid level as inputs, and the water intake flow rate as the decision variable, establish a multi-objective optimization model. The objective function includes tidal energy-saving weights, liquid level interval penalties, and flow smoothing terms. While ensuring the liquid level operates within a safe range, the clear water tank storage window is expanded during high tidal periods, and the energy storage compensation mechanism is activated during low tidal periods. S4: Under the constraints of the above multi-objective optimization model, the optimal water intake flow rate is obtained in real time; S5: Send the optimal water intake flow rate to the water intake pump for execution, and feed back the flow rate and pressure data to the prediction model of S1.
[0006] Furthermore, in the prediction model S1, the water supply time series data Q(t) is decomposed into the following by STL time series decomposition:
[0007] For trend items The ARIMA model is used for prediction, with the input variables being scheduling pressure and temperature; For seasonal items The XGBoost model is used for prediction, and the input features include: Time index: such as the i-th hour of a day, the j-th day of a year, Environmental parameters: temperature, scheduling pressure, Date type: binary code, 0 for weekdays, 1 for public holidays; For residual terms A rule-based corrector is used: when a holiday or emergency occurs, the mean of the residuals from historical holidays or emergencies is superimposed. Using dynamic weighted integration, the predicted water supply for the next 24 hours is output. The scheduling cycle is 5 to 15 minutes, and the system updates and executes the pump group operation control strategy once per cycle.
[0008] Furthermore, the instantaneous water volume calculation model in S1 uses a fuzzy control algorithm to calculate the instantaneous flow rate at the next moment based on the pressure target, pressure change, and flow rate change. The first input variable is E, which is the deviation between the actual pressure and the target pressure. The second input variable is EC, which is the rate of change of the deviation between the actual pressure and the pressure at the previous moment. Output variable: y, which is the instantaneous correction amount that needs to be added to the current flow in the next moment; The rules are arranged based on the principle that large deviations should be corrected significantly and rapid changes in deviations should be slowed down in advance. The reasoning method is the Mamdani method, and the weighted average method is used to resolve fuzziness and calculate the instantaneous flow rate at the next moment.
[0009] Furthermore, the dynamic liquid level safety range in S2 [ , Set and issued by the dispatcher, or automatically generated by connecting to other models, this interval is updated hourly, and its data transmission and interaction support OPC / Modbus protocol and API interface; absolute safety red line. , Set in the PLC safety chain.
[0010] Furthermore, in the multi-objective optimization model of S3, the decision variables are... The optimization function is:
[0011] in: Let t be the amount of water required at time t; This represents the amount of water required at time t-1. Target liquid level; The weight of the liquid level range penalty function is used to prioritize ensuring that the liquid level remains within the safe range. This refers to the tidal energy-saving weighting coefficient. For traffic smoothing weighting coefficients; For quadratic penalty functions that fall outside the safe liquid level range:
[0012] If the liquid level is within the hourly safe range, this item is 0; exceeding the limit will result in a secondary penalty.
[0013] Furthermore, the tidal energy-saving weighting coefficient satisfy: ,
[0014] in, The target liquid level in the clear water tank is determined by taking the median value of the hourly safe liquid level range; Real-time tide level; Total pipeline losses; The overall efficiency of the water intake pump system under current operating conditions; when the tidal level is higher... The smaller, The smaller the value, the more the objective function will incentivize an increase in water intake.
[0015] Furthermore, the target liquid level:
[0016] in, Let be the amount of water required at time t. Predict the water supply at time t. For the target liquid level, Current liquid level The bottom area of the clear water pool For time step, To obtain the supply ratio.
[0017] Furthermore, the optimization model also includes constraints: Liquid level safety red line:
[0018] This refers to the permissible liquid level limit in the clear water tank, i.e., the safety red line for the liquid level. Water intake capacity:
[0019] The influent flow limit is derived by comprehensively considering the operating load of the water plant and the load of the intake pump set.
[0020] Another objective of this application is to provide an energy-saving control system for coordinated water supply and distribution scheduling in a water plant. This system executes the aforementioned energy-saving control method for coordinated water supply and distribution scheduling in a water plant. The system includes: Multi-source data acquisition layer: used to collect panoramic data on the operation of water plants and pipeline networks in real time, including water level gauges of external river water and clear water tanks, flow meters and pressure gauges at the water supply end, motor operation signals and frequencies, as well as weather data, and supports multiple industrial communication protocols such as OPC and Modbus; Data storage management layer: Used to receive, store and manage data uploaded by the multi-source data acquisition layer; it adopts an architecture that combines real-time database and non-relational database. The real-time database is used to store the acquired data, and the non-relational database is used to cache data that is frequently queried but not frequently changed, thereby reducing the pressure on the real-time database and improving the system response speed.
[0021] Intelligent Analysis Layer: The algorithm server deploys the water supply prediction model and the tidal coordinated water intake allocation model. It interacts with the first communication interface in real time with sensor data through the OPC UA / Modbus TCP protocol, and receives scheduling instructions and weather information from external systems through the API interface. Collaborative Execution Layer: The PLC controller establishes a communication connection with the algorithm server of the intelligent analysis layer through the second communication interface, and outputs frequency conversion commands to the water intake pump group, providing real-time feedback on the pump group's operating status to the intelligent analysis layer.
[0022] Furthermore, the PLC controller is connected to the level switch via a hard-wired safety circuit. When the level reaches the absolute safety threshold... , When [the system] stops, it bypasses the software logic and directly triggers an emergency stop of the pump unit; the intelligent analysis layer sends a water intake flow command to the PLC every 5-15 minutes. The PLC dynamically adjusts the output frequency of the frequency converter; the system supports automatic switching of dynamic liquid level safety range during holidays and maintenance modes.
[0023] In summary, this application includes at least one of the following beneficial technical effects: 1. By integrating tidal level, water supply prediction, and clear water tank level constraints into a real-time rolling optimization model, the goal of improving prediction accuracy and maximizing the utilization of tidal potential energy for energy saving and consumption reduction can be achieved.
[0024] 2. The water supply model adopts the STL-ARIMA-XGBoost combined framework. STL decomposes and physically isolates periodic disturbances, ARIMA / XGBoost performs sub-item modeling to make up for the defects of a single model, residual rule base compensates for low-frequency events, and dynamic weights realize scene adaptation. It establishes irreplaceable advantages in terms of error rate, speed, and response to sudden events.
[0025] 3. The water intake allocation model maximizes the use of tidal potential energy and reduces the power consumption of the water intake pumping station under the premise of ensuring the absolute safety of the clear water tank level (no overflow or emptying). It realizes the coordinated scheduling of water supply and water distribution of the water plant's water delivery pumping station, ensures water supply safety and operational efficiency, and provides core decision support for energy-saving scheduling of the water plant. Attached Figure Description
[0026] Figure 1 This is a flowchart of the application; Figure 2 This is a system architecture block diagram of this application. Detailed Implementation
[0027] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application. Furthermore, the technical features involved in the various embodiments described below can be combined with each other as long as they do not conflict with each other.
[0028] like Figure 1 and Figure 2 As shown, the energy-saving control method for coordinated scheduling of water supply and distribution in water plants includes: S1: Construct a mixed prediction model for water supply, including a prediction model and an instantaneous water volume calculation model.
[0029] Predictive Model: Based on historical flow and pressure data, weather data, real-time dispatch instructions, and emergencies at the water supply end, a combined predictive model is used to generate predictions of water supply volume for future periods.
[0030] The water supply time series data Q(t) is decomposed into the following using STL time series decomposition:
[0031] right The ARIMA model is used for prediction, with the input variables being scheduling pressure and temperature; right The XGBoost model is used for prediction, and the input features include: Time index: such as the i-th hour of a day, the j-th day of a year, Environmental parameters: temperature, scheduling pressure, Date type: binary code, 0 for weekdays, 1 for public holidays; right A rule-based corrector is used: when a holiday or emergency occurs, the mean of the residuals from historical holidays or emergencies is superimposed. Using dynamic weighted integration, the predicted water supply for the next 24 hours is output. The scheduling cycle is 5 to 15 minutes, and the system updates and executes the pump group operation control strategy once per cycle.
[0032] Instantaneous water volume calculation model: Combining pressure commands and real-time data, the target instantaneous water supply demand is calculated. Using a fuzzy control algorithm, based on the pressure target, pressure changes, and flow rate changes, the instantaneous flow rate at the next moment is calculated. The first input variable is E, which is the deviation between the actual pressure and the target pressure. The second input variable is EC, which is the rate of change of the deviation between the actual pressure and the pressure at the previous moment. Output variable: y, which is the instantaneous correction amount that needs to be added to the current flow in the next moment; The rules are arranged based on the principle that large deviations should be corrected significantly and rapid changes in deviations should be slowed down in advance. The reasoning method is the Mamdani method, and the weighted average method is used to resolve fuzziness and calculate the instantaneous flow rate at the next moment.
[0033] S2: Obtain the dynamic liquid level safety range and absolute safety red line of the clear water tank. Dynamic liquid level safety range [ , Set and issued by the dispatcher, or automatically generated by connecting to other models, this interval is updated hourly, and its data transmission and interaction support OPC / Modbus protocol and API interface; absolute safety red line. , Set in the PLC safety chain.
[0034] S3: Construct a tidal coordinated water intake allocation model. Using the instantaneous and future demand of S1, the liquid level constraint of S2, the real-time tidal liquid level, and the clear water tank level as inputs, and the water intake flow rate as the decision variable, a multi-objective optimization model is established. The objective function includes tidal energy-saving weights, liquid level interval penalties, and flow smoothing terms. While ensuring the liquid level operates within a safe range, the clear water tank storage window is expanded during high tidal periods, and the energy storage compensation mechanism is activated during low tidal periods.
[0035] Multi-objective optimization model, decision variables are The optimization function is:
[0036] in: Let t be the amount of water required at time t; This represents the amount of water required at time t-1. Target liquid level:
[0037] in, Let be the amount of water required at time t. Predict the water supply at time t. For the target liquid level, Current liquid level The bottom area of the clear water pool For time step, To obtain the supply ratio; The weight of the liquid level range penalty function is used to prioritize ensuring that the liquid level remains within the safe range. The tidal energy-saving weighting coefficient satisfies: ,
[0038] in, The target liquid level in the clear water tank is determined by taking the median value of the hourly safe liquid level range; Real-time tide level; Total pipeline losses; The overall efficiency of the water intake pump system under current operating conditions; when the tidal level is higher... The smaller, The smaller the value, the more the objective function incentivizes an increase in water intake; For traffic smoothing weighting coefficients; For quadratic penalty functions that fall outside the safe liquid level range:
[0039] If the liquid level is within the hourly safe range, this item is 0; exceeding the limit will result in a secondary penalty.
[0040] The optimization model also includes constraints: Liquid level safety red line:
[0041] This refers to the permissible liquid level limit in the clear water tank, i.e., the safety red line for the liquid level. Water intake capacity:
[0042] The influent flow limit is derived by comprehensively considering the operating load of the water plant and the load of the intake pump set.
[0043] S4: Under the constraints of the above multi-objective optimization model, the optimal water intake flow rate is obtained in real time.
[0044] S5: Send the optimal water intake flow rate to the water intake pump for execution, and feed back the flow rate and pressure data to the prediction model of S1.
[0045] This embodiment also provides an energy-saving control system for coordinated water supply and distribution scheduling in water plants. This system executes the aforementioned energy-saving control method for coordinated water supply and distribution scheduling in water plants. The system includes a multi-source data acquisition layer, an industrial gateway, an intelligent analysis layer, and a collaborative execution layer. The system supports automatic switching of dynamic liquid level safety ranges during holidays and maintenance modes.
[0046] Multi-source data acquisition layer: used to collect panoramic data on the operation of water plants and pipeline networks in real time, including water level gauges of external river water and clear water tanks, flow meters and pressure gauges at the water supply end, motor operation signals and frequencies, as well as weather data, and supports multiple industrial communication protocols such as OPC and Modbus; Data storage management layer: Used to receive, store and manage data uploaded by the multi-source data acquisition layer; it adopts an architecture that combines real-time database and non-relational database. The real-time database is used to store the acquired data, and the non-relational database is used to cache data that is frequently queried but not frequently changed (such as dynamic liquid level ranges), so as to reduce the pressure on the real-time database and improve the system response speed.
[0047] Intelligent Analysis Layer: The algorithm server deploys the water supply prediction model and the tidal coordinated water intake allocation model. It interacts with the first communication interface in real time with sensor data through the OPC UA / Modbus TCP protocol, and receives scheduling instructions and weather information from external systems through the API interface. Collaborative Execution Layer: The PLC controller establishes a communication connection with the algorithm server of the intelligent analysis layer through the second communication interface, and outputs frequency conversion commands to the water intake pump group, providing real-time feedback on the pump group's operating status to the intelligent analysis layer. The PLC controller is connected to the level switch via a hard-wired safety circuit; when the level reaches the absolute safety red line... When [the system] stops, it bypasses the software logic and directly triggers an emergency stop of the pump unit. The intelligent analysis layer sends a water intake flow command to the PLC of the system execution layer every 5-15 minutes. The PLC dynamically adjusts the output frequency of the frequency converter.
[0048] like Figure 2 As shown, the multi-source data acquisition layer transmits the collected liquid level, flow rate, and pressure data to the algorithm server via a communication interface. Simultaneously, scheduling commands and weather information are transmitted to the algorithm server via an API interface. The algorithm server uses the above data to solve for the optimal water intake flow rate using a water supply prediction model and a tidal coordinated water intake allocation model. The optimal water intake flow rate is then sent to the water intake pump via a PLC controller for execution, and the flow rate and pressure data are fed back to the prediction model in S1. The algorithm server sends a water intake flow rate command to the PLC in the system execution layer every 5–15 minutes. The PLC dynamically adjusts the inverter's output frequency. The PLC controller is connected to the level switch via a hard-wired safety circuit; when the level reaches the absolute safety red line... In this case, the pump unit is directly triggered to stop, bypassing the software logic. The data storage management layer is used to receive, store, and manage data uploaded by the multi-source data acquisition layer.
[0049] Embodiments of this application may be provided as methods, systems, or computer program products. Therefore, this application may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0050] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1A device that provides the functions specified in one or more boxes.
[0051] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0052] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0053] Content not described in detail in this application specification belongs to the prior art known to those skilled in the art. It is hereby indicated that the above description is intended to help those skilled in the art understand this application, but does not limit the scope of protection of this application. Any equivalent substitutions, modifications, improvements, or simplifications of the above descriptions that do not depart from the essential content of this application fall within the scope of protection of this application.
Claims
1. An energy-saving control method for co-scheduling of water supply and water distribution in a water plant, characterized in that, Comprise: S1: Constructing water supply quantity mixed prediction model, comprising: Prediction model, based on water supply end historical flow, pressure, weather data, real-time scheduling instruction and emergency, through combination prediction model to generate future period water supply quantity prediction result; Instantaneous water quantity calculation model, combining pressure instruction and real-time data, calculating target instantaneous water supply demand; S2: Obtaining clean water pool dynamic liquid level safety interval and absolute safety red line; S3: Constructing tide coordinated water intake quantity distribution model, taking the instantaneous demand and future demand of S1, the liquid level constraint of S2, real-time tide liquid level and clean water pool liquid level as input, taking water intake flow as decision variable to establish multi-objective optimization model, target function includes tide energy-saving weight, liquid level interval penalty and flow smoothing term, while ensuring that the liquid level runs in the safe operation interval, expanding the clean water pool water storage window in the high tide liquid level period, and starting the energy storage compensation mechanism in the low tide liquid level period; S4: Real-time solving the optimal water intake flow under the constraint of the above multi-objective optimization model; S5: The optimal water intake flow is fed to the water intake pump for execution, and the flow and pressure data are fed back to the prediction model of S1.
2. The energy-saving control method for waterworks supply-distribution co-scheduling according to claim 1, characterized in that, In S1, the prediction model decomposes the water supply quantity time series data Q(t) into: ； For the trend term Using ARIMA model, input variables are dispatch pressure and temperature. On the seasonal term For the XGBoost model prediction, the input features include: Time index: such as the i-th hour of the day, the j-th day of the year, Environmental parameters: temperature, scheduling pressure, Date type: binary coding, 0 for weekday, 1 for holiday; For the residual term With a rule corrector: when in a holiday or an event occurs, superimpose the historical holiday or event residual mean; Adopting dynamic weighted integration, outputting prediction water supply quantity in future 24 hours; The scheduling period is 5-15 minutes, and the system updates the pump group operation control strategy and executes once every period.
3. The energy-saving control method for waterworks supply-distribution co-scheduling according to claim 1, characterized in that: The instantaneous water quantity calculation model in S1 is calculated by a fuzzy control algorithm based on pressure target, pressure change and flow change, wherein, First input variable: E, the deviation of actual pressure and target pressure; Second input variable: EC, the deviation change rate of actual pressure and last time pressure; Output variable: y, instantaneous correction amount needed to be superimposed on current flow at next time; Based on rule base principle arrangement, the principle is that large deviation is greatly corrected and fast deviation change is slowed down in advance, the reasoning method is Mamdani method, and weighted average method is used to solve the ambiguity, and the next time instantaneous flow is calculated.
4. The energy-saving control method for water-plant supply-distribution co-scheduling according to claim 1, characterized in that: Dynamic liquid level safety interval in S2 , ] is set and issued by the dispatcher, or automatically generated by accessing other models. The interval is updated every hour, and its data transmission and interaction support OPC / Modbus protocol and API interface; Absolute safety red line[ , ] is set in the PLC safety chain.
5. The energy-saving control method for waterworks supply-distribution co-scheduling according to claim 1, characterized in that, In the multi-objective optimization model in S3, the decision variable is , and the optimization function is: ； Wherein: Qt is the required water quantity at time t; Qt-1 = required water quantity at time t-1; Target level; The liquid level interval penalty function weight is used to ensure that the liquid level is in the safe interval. is the tidal energy saving weight coefficient; is a flow smoothing weight factor; For the quadratic penalty function falling outside the liquid level safety interval range: ； The liquid level is within the hourly liquid level safety interval, and this item is 0, and a secondary penalty is generated when it is out of range.
6. The energy-saving control method for water-plant supply-distribution co-scheduling according to claim 5, characterized in that, Tidal energy weight coefficient satisfies: ， ； wherein, is the target water level of the clear water tank, and the median value of the short-term water level safety interval is taken; is the real-time tidal level; is the total loss of the pipeline; is the total efficiency of the water pump system under the current working condition; when the tidal level is higher, is smaller, is smaller, and the target function incentive increases the water intake.
7. The energy-saving control method for waterworks supply-distribution co-scheduling according to claim 5, characterized in that, Target liquid level: ； wherein, is the required water intake at time t, is the predicted water supply at time t, is the target liquid level, is the current liquid level, is the clean water tank bottom area, is the time step, is the intake-to-supply ratio.
8. The energy-saving control method for water-plant supply-distribution co-scheduling according to claim 5, characterized in that, The optimization model also includes constraint conditions: Liquid level safety red line: ； the permissible level limit for the water tank, i.e. the level safety red line; Inlet capacity: ； The water intake flow limit is obtained by comprehensively considering the operation load of the water plant and the load of the water intake pump set.
9. An energy-saving control system for co-scheduling of water supply and water distribution in a water plant, characterized in that, The system executes the energy-saving control method of water plant water supply-distribution collaborative scheduling according to any one of claims 1-7, comprising: Multi-source data acquisition layer: used for real-time acquisition of panoramic data of water plant and pipe network operation, including outer river water and clean water pool liquid level meter, water supply end flow meter and pressure gauge, motor operation signal and frequency, and weather data, etc., and supports multiple industrial communication protocols such as OPC and Modbus; Data storage management layer: used for receiving, storing and managing data uploaded by multi-source data acquisition layer; it adopts the architecture combining real-time database and non-relational database, the real-time database is used for storing acquisition data, and the non-relational database is used for caching frequently queried and infrequently changed data, reducing the pressure of real-time database and improving the response speed of system; Intelligent analysis layer: algorithm server of water supply prediction model and tidal coordinated water intake allocation model is deployed, real-time interaction of sensing data is realized through OPCUA / Modbus TCP protocol and first communication interface, and dispatching instructions and weather information are received from external system through API interface; Coordinated execution layer: PLC controller establishes communication connection with algorithm server of intelligent analysis layer through second communication interface, and outputs frequency conversion instruction to water intake pump group, and real-time feedback of pump group running state is realized to intelligent analysis layer.
10. The energy-saving control system for the coordinated scheduling of water supply and water distribution of a water plant according to claim 9, characterized in that: The PLC controller is connected to the level switch via a hard-wired safety circuit. When the level reaches the absolute safety red line... , When [the system] stops, it bypasses the software logic and directly triggers an emergency stop of the pump unit; the intelligent analysis layer sends a water intake flow command to the PLC every 5-15 minutes. The PLC dynamically adjusts the output frequency of the frequency converter; the system supports automatic switching of dynamic liquid level safety range during holidays and maintenance modes.