Urban water supply pipe network intelligent scheduling method and system based on internet of things cloud platform

Abstract

The application discloses a city water supply pipe network intelligent scheduling method and system based on an internet of things cloud platform, and relates to the technical field of water cloud management. The method divides the city water supply pipe network into a plurality of water monitoring units, and connects the data of the water monitoring units to a One NET internet of things cloud platform through NB-IoT, constructs internal state variables of the water monitoring units and verifies the effectiveness. In each water monitoring period, the flow and pressure changes of the water monitoring units are analyzed to determine whether the local water supply meets the predicted demand. If the predicted demand is not met, multi-step prediction calculation for water supply scheduling is performed, and the valve opening and / or pump station start-stop timing is adjusted in stages through the downlink of NB-IoT. If the predicted demand is met, short-term water demand prediction is performed, and a steady-state confirmation prediction process of the water use trend is performed. The method realizes fine and closed-loop intelligent scheduling of the city water supply pipe network.

Description

Technical Field

[0001] This invention relates to the field of water cloud management technology, and in particular to a method and system for intelligent scheduling of urban water supply networks based on an Internet of Things (IoT) cloud platform. Background Technology

[0002] Narrowband Internet of Things (NB-IoT) communication technology is a low-power wide-area network technology designed for massive IoT applications, offering advantages such as deep coverage, low power consumption, large connectivity, and high reliability. In smart water management systems, NB-IoT terminals can be deployed in water supply networks, water meters, water quality monitoring points, and key equipment to enable long-term, low-power data collection of flow, pressure, water quality, and equipment status. This data is then uploaded to a cloud platform via cellular networks for data storage, remote monitoring, and basic analysis.

[0003] Existing technologies deploy NB-IoT terminal nodes in water supply networks, water meters, water quality monitoring points, and key equipment, integrating sensors for flow, pressure, water quality, and equipment status. Collected data is uploaded to a cloud platform via the NB-IoT cellular network, leveraging its deep coverage, massive connectivity, and high reliability to ensure data transmission stability and terminal energy efficiency. Some systems configure edge computing modules in the IoT units to fuse, compress, and preprocess data for anomalies. The cloud platform centrally stores, structures, and performs basic statistical analysis on the data, generating flow distribution maps, water quality monitoring curves, and equipment operating status tables. Based on cloud data, the system can optimize water supply scheduling, analyze leakage, predict equipment failures, and issue water quality warnings. Real-time operating status is displayed to management personnel through a visual interface, enabling remote monitoring and centralized control. Finally, scheduling or maintenance commands generated by the cloud platform are sent to terminals via the NB-IoT downlink, automatically adjusting valve and pump station operation or issuing alarm notifications, forming a closed-loop smart water management system.

[0004] However, with the acceleration of urbanization, urban water management faces increasing challenges, such as water supply security, water quality assurance, and pipeline leakage. Especially during peak summer water consumption periods, water demand in some areas increases dramatically in a short period. Existing forecasting methods, based solely on historical average data, cannot accurately reflect actual demand, resulting in delayed water supply dispatch responses. This manifests as fluctuations in pipeline pressure, frequent pump station start-ups and shutdowns, increased energy consumption and leakage, and insufficient water pressure for some high-rise users, thus affecting residents' normal water usage experience and the stable operation of the water supply system. Therefore, improving the accuracy of hourly water demand forecasting and achieving closed-loop linkage between forecast results and dispatch control are urgent technical problems that need to be solved in current urban smart water management. Summary of the Invention

[0005] Therefore, embodiments of the present invention provide a method and system for intelligent scheduling of urban water supply networks based on an Internet of Things (IoT) cloud platform. The technical solution is as follows:

[0006] On the one hand, a smart scheduling method for urban water supply networks based on an IoT cloud platform is provided. This method includes: dividing the urban water supply network into several water monitoring units; connecting the water usage data and equipment operation data of each water monitoring unit to the One NET IoT cloud platform via NB-IoT; constructing internal state variables for each water monitoring unit and verifying their effectiveness; the internal state variables include at least instantaneous flow rate, cumulative water volume, network pressure value, and pump and valve status; within each water monitoring cycle, by analyzing the changes in flow rate and pressure of each water monitoring unit, determining whether the local water supply meets the predicted demand; if it is determined that the local water supply does not meet the predicted demand, then performing multi-step prediction calculations for water supply scheduling, and adjusting valve opening and / or pump station start-up and shutdown sequence in stages through the downlink of NB-IoT; if it is determined that the local water supply meets the predicted demand, then predicting the short-term water demand within the current water monitoring cycle, and performing a steady-state confirmation prediction process for water usage trends.

[0007] On the other hand, an intelligent scheduling system for urban water supply networks based on an IoT cloud platform is provided. This system includes: a monitoring unit construction and validity verification module, which divides the urban water supply network into several water monitoring units, connects the water usage data and equipment operation data of each water monitoring unit to the One NET IoT cloud platform via NB-IoT, constructs the internal state variables of the water monitoring units and verifies their validity; a local water supply determination module, which determines whether the local water supply meets the predicted demand by analyzing the flow and pressure changes of each water monitoring unit in each water monitoring cycle; if the local water supply does not meet the predicted demand, it enters the multi-step scheduling management module; if the local water supply meets the predicted demand, it enters the steady-state trend confirmation module; the multi-step scheduling management module is used to perform multi-step prediction calculations for water supply scheduling and adjusts the valve opening and / or pump station start-up and shutdown sequence in stages through the downlink of NB-IoT; and the steady-state trend confirmation module is used to predict the short-term water demand in the current water monitoring cycle and execute the steady-state confirmation prediction process for water usage trends.

[0008] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0009] 1. This invention addresses the problems of delayed dispatch response and increased energy consumption caused by insufficient prediction accuracy in existing urban water management. Firstly, the urban water supply network is divided into several water monitoring units. Real-time data collection of internal state variables such as instantaneous flow rate, cumulative water volume, network pressure, and pump / valve status is achieved via NB-IoT and connected to the OneNET platform for validity verification, constructing a comprehensive monitoring system. Within each monitoring cycle, the flow and pressure changes of each unit are analyzed to determine whether the local water supply meets the predicted demand. If not, multi-step prediction calculations for water supply dispatch are performed, and valve opening and pump station start-up / shutdown timing are adjusted in stages via the NB-IoT downlink to achieve closed-loop linkage of perception, prediction, and control. If the demand is met, short-term water demand is predicted, and steady-state confirmation of water usage trends is performed. This method effectively improves the accuracy of hourly water demand prediction through refined monitoring and scenario-based prediction, solves the problem of dispatch lag during peak water usage periods, reduces network pressure fluctuations and frequent pump station start-ups and shutdowns, and ensures stable water pressure for high-rise users, achieving deep integration of prediction results and dispatch control.

[0010] 2. This invention further refines the construction and validity verification method of internal state variables of the water monitoring unit, aiming to improve the accuracy of data acquisition and the stability of pressure regulation. First, based on basic water supply attribute information, state parameters are constructed for each monitoring element, forming internal state variables such as instantaneous flow rate and cumulative water volume. Simultaneously, under the constraint of a preset flow rate change step size, a metering consistency constraint is established between the cumulative water volume and the instantaneous flow rate, ensuring that the integral results within the same period match, thus guaranteeing data reliability from the source and avoiding misjudgments and misadjustments caused by metering deviations. Based on this, combined with the pressure characteristics of the pipeline network nodes, the pressure fluctuation range, benchmark threshold, and response time limit are determined to quantitatively characterize the degree of pressure stability and to determine in real time whether there is a first or second unstable fluctuation. When the first unstable fluctuation is detected, the first management strategy is immediately triggered, controlling the pump valves to gradually decrease the flow rate according to a preset step size through a flow suppression adjustment command to prevent a sudden pressure surge from impacting the pipeline network. When the second unstable fluctuation is detected, the second management strategy is executed, issuing a pressure compensation adjustment command to gradually increase the flow rate to quickly respond to underpressure demands. This process employs a graded response strategy for unstable fluctuations, enabling refined control by proactively reducing pressure during overpressure and rapidly compensating for underpressure. This reduces sudden changes in pipeline pressure and frequent start-ups and shutdowns of pump stations, effectively preventing insufficient water pressure for high-rise users, especially during peak water usage periods.

[0011] 3. This invention establishes a precise judgment and response mechanism for determining whether local water supply meets predicted demand. First, within each water monitoring cycle, the instantaneous flow sequence of the local water supply area is cumulatively summed to obtain the cumulative flow value. Simultaneously, the pressure change rate is obtained by time differentiation of the pipeline pressure sequence. By periodically integrating or cumulatively calculating both, a water supply judgment parameter is generated that quantifies the degree of matching between local water supply capacity and preset demand. This judgment method integrates flow accumulation and pressure change rate analysis, enabling more sensitive detection of dynamic imbalances in regional water supply and helping to identify potential local water supply insufficiency in advance.

[0012] 4. If the water supply judgment parameters exceed the preset water supply judgment parameters, it is determined that the local water supply does not meet the predicted demand, and multi-step prediction calculations for water supply scheduling are executed. By fitting the instantaneous flow rate, cumulative water volume, and pipeline pressure values ​​in a time series, a trend function reflecting the operating law of each monitoring unit is determined. Then, the water supply volume and pressure adjustment amount required to meet the short-term water supply are calculated, and the valve opening adjustment range and pump station start-up and shutdown sequence are generated. After the phased adjustment plan is formed, it is constrained and verified, including whether the pump station operating power exceeds the limit and whether the valve opening change step size meets the requirements. The adjusted pipeline pressure is simulated to ensure that it is within the safe range. If there are any over-limit items, personnel are prompted to make corrections; otherwise, scheduling commands are output through the NB-IoT downlink to execute phased flow regulation and pressure control. This process ensures that the scheduling plan meets short-term needs while remaining within the equipment's capacity, avoiding pump station overload or sudden changes in pipeline pressure caused by improper instructions; phased adjustment and NB-IoT distribution reduce pressure fluctuations and frequent pump station starts and stops, while ensuring stable water pressure for high-rise users and improving the safety of the water supply system.

[0013] 5. If the water supply determination parameters do not exceed the preset water supply determination parameters, the local water supply is determined to meet the predicted demand, and the process transitions to short-term water demand prediction and steady-state confirmation of water use trends. First, the stored time-series prediction model is invoked, and combined with real-time water use data and operational data of various monitoring elements under the current water monitoring scenario, the model is adaptively updated to form a water demand prediction model suitable for the current scenario. Then, using data such as instantaneous flow rate, cumulative water volume, pipeline pressure, valve opening, pump station power, and start / stop status of each water monitoring unit as input, the model performs time-series calculations and outputs the short-term water demand prediction value for the current monitoring period. This process achieves dynamic coupling between the prediction model and real-time operating conditions, making the prediction results more consistent with actual water use patterns. After obtaining the predicted values ​​for a consecutive preset number of monitoring periods, they are time-series sorted and smoothed to generate a water demand change trend curve, and the predicted water demand change between adjacent periods is calculated to obtain the predicted water demand change trend. Further linear fitting analysis was performed on its changing trend to calculate the trend slope. Based on the defined range of slope values, this slope was mapped to a trend discrimination parameter: a slope greater than zero resulted in a parameter of 1, indicating an upward trend in water demand; a slope equal to zero resulted in a parameter of 0, indicating a steady state; and a slope less than zero resulted in a parameter of -1, indicating a downward trend. This quantitative mapping allows for a direct identification of the dynamic direction of water demand. This prediction and trend analysis mechanism forms a closed loop of prediction, verification, and optimization, further enhancing the intelligence level of water management and ensuring the stable operation of the water supply system. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 A flowchart of an intelligent scheduling method for urban water supply networks based on an Internet of Things cloud platform, provided in an embodiment of the present invention;

[0016] Figure 2 This is an overall block diagram of a smart water meter system provided in an embodiment of the present invention;

[0017] Figure 3 This is a hardware structure diagram of a smart water meter system terminal provided in an embodiment of the present invention;

[0018] Figure 4 A flowchart illustrating the device access process of the One NET IoT cloud platform provided in this embodiment of the invention;

[0019] Figure 5 This is a functional structure diagram of the smart water management platform provided in an embodiment of the present invention;

[0020] Figure 6 This is a diagram illustrating the overall architecture of the smart water management platform provided in an embodiment of the present invention.

[0021] Figure 7 A diagram illustrating the architecture of the water demand prediction model provided in this embodiment of the invention;

[0022] Figure 8 This is a schematic diagram of the structure of an intelligent dispatching system for urban water supply networks based on an Internet of Things cloud platform, provided in an embodiment of the present invention. Detailed Implementation

[0023] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0024] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0025] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0026] Smart water meter systems based on NB-IoT communication technology and the One NET IoT cloud platform, such as Figure 2 The diagram shows the overall structure of the smart water meter system. This system mainly comprises three subsystems: the water meter terminal, the One NET cloud platform, and the water meter monitoring platform. The water meter terminal integrates data acquisition and cloud interaction functions, responsible for accurately collecting water flow information and transmitting it to the cloud via an NB-IoT module. The One NET cloud platform, acting as a crucial hub between the device and the monitoring backend, greatly simplifies the development process of the monitoring system and reduces the complexity of subsequent maintenance through its API interface and real-time data push mechanism. The water meter monitoring platform comprehensively receives and processes diverse data from the cloud platform, completes interaction with the meter, and achieves a comprehensive integrated service encompassing remote device control, automatic data collection and reporting, real-time monitoring, data storage, and visualization.

[0027] like Figure 3The diagram shows the terminal hardware structure of the smart water meter system. This design uses the STM32L151RCT6 low-power microcontroller module (MCU) as the core component. It integrates a serial port, crystal oscillator, and GPIO sub-modules, and connects to the pulse detection module, communication module, power supply module, valve control module, and storage module. The pulse detection module collects water flow data. The NB-IoT communication module integrates a communication antenna and a SIM card reader, enabling network access and data exchange via AT command sets. The valve control module executes valve opening and closing commands. The storage module stores critical data. The power supply module provides stable power to the MCU, NB-IoT communication module, and valve control unit after voltage regulation by a voltage regulator chip.

[0028] like Figure 4 The diagram shown illustrates the device access process for the One NET IoT cloud platform. The device access process includes six steps: user registration, product creation, device creation, hardware access, online deployment, and application development. To achieve ultra-low power consumption and system stability, the NB-IoT module uses the Lightweight Machine-to-Machine (LWM2M) protocol to communicate with the IoT platform. LWM2M defines three logical entities: the LWM2M Server, which is the access machine and acts as the interface to the platform server; the Bootstrap Server, which is responsible for configuring the LWM2M client; and the LWM2M Client, which executes commands issued by the server and reports the execution results.

[0029] like Figure 5 The diagram shows the functional structure of the smart water management platform. The platform provides differentiated functional modules for different user categories: super users can access remote meter reading services and user account management services; water meter plant administrators are responsible for water usage information management services and payment management services; waterworks administrators use revenue query services and report management services; and ordinary users enjoy file management services, forming a multi-level functional system covering the entire business process.

[0030] like Figure 6The diagram shows the overall architecture of the smart water management platform. The platform adopts a layered architecture design, from bottom to top: the perception layer, including NB-IoT smart water meters, water quality monitoring, water pressure detection, flow meters, and other equipment; the infrastructure layer, including hardware support such as hosts, networks, and storage; the platform layer, including data centers, water management platforms, water GIS support platforms, monitoring platforms, and water consumption data analysis platforms; the application layer, including functions such as remote meter reading, water management systems, GIS systems, pipeline leakage detection, scheduling optimization, automatic billing, anomaly alarms, and water consumption prediction; and the presentation layer, including browser access, mobile APP, WeChat official account, and unified portal, achieving full-chain coverage from data collection to business applications.

[0031] Based on the complete technical carrier comprised of the aforementioned smart water meter terminal, One NET IoT cloud platform, and smart water management platform, NB-IoT narrowband IoT technology enables data connectivity between the sensing layer devices and the platform layer, constructing a smart water management infrastructure system covering the entire chain of data acquisition, network transmission, cloud platform processing, and business applications. On this basis, this invention provides a method for intelligent scheduling of urban water supply networks based on an IoT cloud platform, such as... Figure 1 The flowchart shown is for an intelligent scheduling method for urban water supply networks based on an IoT cloud platform. The processing flow of this method may include the following steps:

[0032] Step 1

[0033] First, the urban water supply network is divided into several water monitoring units, each covering a specific network area. Within each unit, monitoring equipment such as flow meters, pressure transmitters, and pump / valve controllers are installed to achieve real-time data collection of water flow, cumulative water consumption, network pressure, and pump / valve operating status. Through NB-IoT communication technology, the water consumption data and equipment operation data collected by each water monitoring unit are stably and with low latency connected to the One NET IoT cloud platform, enabling centralized management, real-time monitoring, and unified storage of data across the entire network.

[0034] Each water monitoring unit is functionally divided, the monitoring scope and key nodes are defined, and several monitoring elements are arranged within the unit, including flow meters, pressure transmitters and pump and valve controllers. The flow meters are used to collect water supply flow information in real time, the pressure transmitters are used to obtain the pressure values ​​of each node in the pipeline network, and the pump and valve controllers are used to record the power of the pump station and the valve opening status, while supporting the execution and feedback of control commands.

[0035] Based on preset basic water supply attribute information, including parameters such as the length, diameter, number of nodes, design flow rate, and rated pressure of each unit network, the reference operating range and tolerance range for each monitoring element are determined. For example, for instantaneous flow rate, the maximum allowable flow rate variation per unit time can be set as the preset flow rate variation step size. This step size can be determined based on the network design flow rate, pump station output capacity, and statistical characteristics of historical operating data. For instance, in residential areas, flow rate fluctuations are typically within ±5%, while in industrial water areas, this can be appropriately increased to ±10%. Under this constraint, a metering consistency constraint is established for the integral results of cumulative water volume and instantaneous flow rate within the same metering period. That is, by comparing the cumulative value of instantaneous flow rate within the period with the actual cumulative water volume, the logical consistency of the measurement data is judged, ensuring that the data collected by the flow meter and cumulative metering equipment does not have abnormal drift or equipment failure. For example, if the instantaneous flow rate is 1200 cubic meters in a certain hour, while the cumulative water volume is 1190 cubic meters, the difference between the two is only 0.83%, which is within the preset metering consistency range, and the metering data can be considered valid; if the difference exceeds the preset range, it is marked as abnormal data and further verification is triggered.

[0036] After completing the metering consistency constraints, the pressure values ​​of the pipeline network are further constructed by combining the pressure characteristics of each pipeline node in the basic water supply attribute information. Specifically, this includes determining the pressure fluctuation range, pressure benchmark threshold, and pressure regulation response time limit. The pressure fluctuation range can be set based on the pipeline network design pressure, historical operating data, and actual user water consumption characteristics. For example, the pressure fluctuation range for residential area nodes is allowed to be 0.3-0.5 MPa, and the pressure fluctuation range for high-rise building nodes is allowed to be 0.4-0.6 MPa. The pressure benchmark threshold is usually set as the average value of the pipeline network design pressure, such as 0.4 MPa for residential areas and 0.5 MPa for high-rise building nodes. The pressure regulation response time limit is set according to the regulation capacity of pump stations and valves, usually 1 to 5 minutes, to measure the response speed of the pipeline network to regulation commands, ensuring that the system can return to a stable state within a reasonable time.

[0037] Taking a specific calculation example, within the monitoring period of a water monitoring unit in a residential area, the standard deviation of the node pressure sequence was calculated to be 0.042 MPa. At the same time, the number of pump and valve controller starts and stops was 2, which is lower than the preset allowable number of starts and stops per hour of 3. Moreover, the flow fluctuation range was within the preset threshold range. Here, the preset threshold range refers to the allowable flow fluctuation range determined based on the pipeline design pressure and historical operating data. For example, the flow fluctuation per unit time shall not exceed ±5% or ±0.02 cubic meters per second. Meanwhile, the number of pump and valve starts and stops is within the maximum allowable number of starts and stops per hour, for example, not exceeding 3 times. Since the pressure standard deviation is less than the preset fluctuation threshold of 0.05 MPa, and the number of pump and valve start-ups and flow fluctuations are all within the allowable range, the pipeline pressure of the water monitoring unit is determined to be in a stable state. The allowable range includes a pressure standard deviation not exceeding 0.05 MPa, a pump and valve start-up and shutdown frequency not exceeding 3 times per hour, and flow fluctuation amplitude within the above preset limits. If the pressure standard deviation exceeds 0.05 MPa or the number of pump and valve start-ups and shutdowns exceeds the preset allowable number of start-ups and shutdowns of 3 times per hour, it indicates that there is abnormal fluctuation in the pipeline pressure, and an unstable fluctuation judgment should be triggered to achieve proactive control and risk warning.

[0038] Secondly, it is necessary to determine whether a first or second unstable fluctuation exists. A first unstable fluctuation indicates that the pipeline pressure is higher than the upper limit of the preset stable range, meaning the pipeline pressure is too high. This could lead to overflow at the user end, pipeline damage, or abnormal valve control. In this case, real-time monitoring of the pressure sequence reveals that the pressure exceeds the upper limit for several consecutive cycles. For example, if the pressure at a residential area node remains above 0.55 MPa for three consecutive minutes, this triggers the first unstable fluctuation determination. A second unstable fluctuation indicates that the pipeline pressure is lower than the lower limit of the preset stable range, meaning the pipeline pressure is insufficient. This could lead to insufficient water pressure for high-rise users or interruption of water supply at the end of the line. In this case, pressure fluctuation sequences reveal that the pressure is lower than the lower limit for several consecutive cycles. For example, if the pressure at a high-rise building node is below 0.45 MPa for five consecutive minutes, this is determined to be a second unstable fluctuation.

[0039] The preset upper and lower limits of the stable range are safety ranges determined based on the pipeline network design pressure and the water supply requirements of the nodes. The upper limit is usually 105% to 110% of the pipeline network design pressure, and the lower limit is usually 90% to 95% of the design pressure, used to ensure pipeline network safety and the water supply needs of end users. For example, for a residential area node design pressure of 0.5 MPa, the upper limit of the stable range can be set at 0.55 MPa and the lower limit at 0.45 MPa; for a high-rise building node design pressure of 0.55 MPa, the upper limit is 0.605 MPa and the lower limit is 0.495 MPa.

[0040] Through the aforementioned construction and judgment process, the internal state variables not only accurately reflect the operating status of the water monitoring unit, but also, combined with preset basic water supply attribute information and historical operating data, achieve quantitative management of pipeline pressure and water supply flow. The resulting stability indicators can provide a scientific basis for subsequent water supply scheduling, valve opening adjustment, and pump station start-up and shutdown control, enabling real-time monitoring, anomaly identification, and proactive adjustment of the urban water supply network, thereby effectively improving the safety, reliability, and operating efficiency of the water supply system.

[0041] When there is no first or second unstable fluctuation within a certain monitoring period, the pipeline pressure, instantaneous flow rate, and pump and valve start-stop status of each water monitoring unit are monitored in real time, and the pressure fluctuation sequence, flow rate change data, and number of pump and valve status changes are recorded for subsequent water supply trend analysis, pipeline pressure stability assessment, and scheduling strategy optimization reference.

[0042] When the first unstable fluctuation is detected, i.e., the pipeline pressure is higher than the upper limit of the preset stable range, the corresponding first management strategy is triggered. This involves issuing a flow suppression and adjustment command to achieve real-time control of the pump and valve controllers. This includes gradually reducing the valve opening and / or reducing the pump station output power, so that the water supply flow decreases step by step according to the preset flow change step size. Each decrease is 50% to 100% of the preset flow change step size. For example, if the preset flow change step size is 2 cubic meters per hour, each decrease can be adjusted within the range of 1 to 2 cubic meters per hour. This effectively suppresses further increases in pipeline pressure and avoids overflow at the user end, pipeline damage, abnormal valve control, or pump station overload operation.

[0043] When a second unstable fluctuation is detected, i.e., the pipeline pressure is lower than the lower limit of the preset stable range, the system triggers the corresponding second management strategy. By issuing pressure compensation adjustment commands, the system performs real-time control of the pump and valve controllers, including gradually increasing the valve opening and / or increasing the pump station output power. This causes the water supply flow to increase step by step according to the preset flow change step size, with each increment being 50% to 100% of the preset flow change step size. For example, if the preset flow change step size is 2 cubic meters per hour, each increment can be adjusted within the range of 1 to 2 cubic meters per hour. This increases the water pressure at the end nodes, ensuring that the water supply needs of high-rise users and key nodes are met, while avoiding pipeline overpressure or pump station power overload.

[0044] Finally, after completing the adjustment operations corresponding to the first or second management strategy, the pipeline pressure, flow rate, and pump / valve status of each monitoring unit are monitored and analyzed again. If no first or second unstable fluctuation occurs after adjustment, routine monitoring and data recording continue, and the sequence data of pressure, flow rate, and pump / valve status are periodically analyzed to provide a basis for scheduling and prediction in the next monitoring cycle. If any unstable fluctuation still exists after adjustment, an unstable fluctuation risk warning will be triggered, abnormal pressure status will be recorded and alarmed, and control suggestions will be provided to operators, including but not limited to further adjusting valve opening, adjusting pump station start-up and shutdown sequence, activating standby pump stations, or temporarily restricting flow in specific areas.

[0045] Through the aforementioned monitoring, regulation, and early warning processes, closed-loop control of urban water supply network pressure is achieved: it can proactively implement flow suppression or pressure compensation regulation when the network pressure is abnormal, avoiding water supply risks caused by overpressure or underpressure; and it can continuously monitor and record operational data when the pressure is stable, providing real-time data support for short-term water demand forecasting and multi-step scheduling. This enables intelligent and dynamic management of the urban water supply network, improves water safety, water supply reliability, and user experience, reduces pump station energy consumption and network losses, and ensures the stable and efficient operation of the urban water system.

[0046] Step Two

[0047] Within each water monitoring cycle, the instantaneous flow rate and pipeline pressure changes of each water monitoring unit are monitored. Based on the pipeline topology, node importance, and regional water usage characteristics, the pipeline area to be assessed and its corresponding monitoring unit set are identified as a local water supply area. The instantaneous flow rate sequence within the local water supply area is cumulatively summed, and the result is recorded as the cumulative flow rate value. Simultaneously, the pipeline pressure sequence within the local water supply area is differentiated over time, and the result is recorded as the pressure change rate. In the above process, the instantaneous flow rate and pipeline pressure of the local water supply area are normalized or de-unitized, and the cumulative flow rate and pressure change rate are dimensionless values, facilitating unified analysis and comparison.

[0048] By performing periodic integration or cumulative calculations on the cumulative flow rate and pressure change rate, water supply judgment parameters are obtained to determine the degree of matching between local water supply capacity and preset demand. The calculation formula is as follows:

[0049] ;

[0050] In the formula, W represents the water supply determination parameter, t0 represents the start time of the current water monitoring cycle, and T represents the length of the water monitoring cycle, i.e., the time interval for integration. This represents the cumulative flow value within the local water supply area, t0+T represents the duration of the current water monitoring cycle, i.e., the total integration time, i represents the monitoring node number within the local water supply area, and R represents the set of all water monitoring units within the local water supply area, including the monitoring node numbers that need to be determined. This represents the flow rate weighting coefficient, used to adjust the contribution weight of flow rate to water supply determination parameters. This represents the pressure fluctuation weighting coefficient, used to adjust the contribution weight of pipeline pressure fluctuations to water supply determination parameters. This represents the pressure change rate at the i-th monitoring node within the local water supply area.

[0051] The setting of flow weighting coefficient and pressure fluctuation weighting coefficient directly affects the sensitivity of water supply judgment parameters to local water supply capacity. Typically, the flow weighting coefficient can be empirically or statistically fitted based on the matching degree between historical flow data and actual water demand, ensuring that cumulative flow plays a dominant role in the judgment parameters to reflect whether the total water supply within the water monitoring cycle meets demand. The pressure fluctuation weighting coefficient, on the other hand, can be adjusted by considering the pipeline design pressure and the impact of pressure fluctuations on equipment safety and user water supply stability, so that pressure changes reflect the risk of water supply fluctuations in the judgment parameters, while avoiding over-amplifying small fluctuations to prevent misjudgments.

[0052] The aforementioned flow-weighted coefficient and pressure fluctuation-weighted coefficient adjust the influence weights of flow and pressure on water supply determination, respectively, so that the calculation results can simultaneously reflect the adequacy of water supply and pressure stability. Secondly, this method can weight the contributions of different monitoring nodes and consider the cumulative effect over time, thereby avoiding misjudgments caused by instantaneous fluctuations and improving the accuracy of water supply determination. Finally, the obtained water supply determination parameters can comprehensively reflect the adequacy and stability of water supply in a local water supply area throughout the entire water monitoring cycle, improving the safety of pipeline network operation and the reliability of water supply.

[0053] If the water supply determination parameters exceed the preset water supply determination parameters, it is determined that the local water supply does not meet the predicted water demand, and the process corresponding to Scenario 1 is executed; otherwise, it is determined that the local water supply meets the predicted water demand, and the process corresponding to Scenario 2 is executed. The preset water supply determination parameters are represented by the summation and averaging of historical water supply determination parameters from the historical urban hourly water demand prediction process.

[0054] Scenario 1: Perform multi-step predictive calculations for water supply scheduling, and adjust valve opening and / or pump station start-up and shutdown timing in stages via NB-IoT downlink, specifically:

[0055] According to a preset time series, the current instantaneous flow rate, cumulative water volume, and pipeline pressure values ​​of each water monitoring unit within the local water supply area are sorted. Based on the sorted data, the operating status of each monitoring unit in the current period is fitted. During the fitting process, multinomial regression, spline curves, or adaptive smoothing algorithms based on historical operating data are used. The observed values ​​at each time point are taken as input, and the optimal fitting curve is obtained by minimizing the error between the fitted curve and the actual observed values, thereby generating the flow trend function and pressure trend function for each water monitoring unit. The horizontal axis of the optimal fitting curve represents the time point corresponding to the collected data, and the vertical axis represents the instantaneous flow rate, cumulative water volume, or pipeline pressure value at each time point. The horizontal and vertical axes are paired to form a continuous time series.

[0056] The fitted curve can continuously reflect the dynamic changes in flow and pressure of a local water supply unit within the current cycle, including the trend of water supply increase or decrease, pressure fluctuation amplitude, and possible periodic changes. Furthermore, by calculating the first derivative of the curve, the rate of change in flow and pressure can be obtained, which can be used to quantify the rate and trend of change during short-term water supply processes. The fitted curve can not only be used to monitor the current water supply status but also serve as a basis for predicting short-term water supply and pressure adjustment needs, supporting phased control of valve opening and pump station start-up and shutdown sequences.

[0057] Based on the trend function, the required water supply and pressure adjustment for each monitoring unit to meet short-term water demand within the current period can be further calculated. Specifically, taking a monitoring unit as an example, the trend function predicts a cumulative water supply of 500 cubic meters and an average pressure of 2.5 MPa within the current period. According to short-term water demand, this unit needs to provide 550 cubic meters of water with a pressure maintained at 2.7 MPa. Comparing the trend function prediction with the demand value, the flow difference is 550 - 500 = 50 cubic meters, and the pressure difference is 2.7 - 2.5 = 0.2 MPa. A positive difference indicates that the water supply or pressure needs to be increased, while a negative difference indicates that the water supply or pressure can be reduced. Similarly, the flow and pressure differences are calculated for each monitoring unit. Subsequently, these differences are combined with the hydraulic characteristics of the pipeline network and pump and valve control parameters to convert them into valve opening adjustment ranges and pump station start / stop or output power adjustment values, forming executable control parameters. This method quantifies the water supply adjustment needs of each node in the current cycle, ensuring that the flow and pressure meet the water demand requirements, while taking into account the safety of the pipeline pressure and the water balance, thus achieving a match between local water supply capacity and demand.

[0058] Subsequently, based on the calculated adjustment amounts, corresponding valve opening adjustment ranges and pump station start-up and shutdown sequences are generated for phased regulation of pipeline flow and pressure. This ensures that the actual water supply and pressure of each monitoring unit can be dynamically adjusted according to the target values ​​calculated by the trend function. The phased regulation plan formed by the valve opening adjustment range and pump station start-up and shutdown sequence needs to be constrained and verified, including pump station operating power limits and / or valve opening change step size limits. The pump station operating power limit is determined by obtaining the current power of the pump station and comparing it with the rated power to ensure that the start-up, shutdown, and output power in the plan do not exceed the upper limit of the pump station's rated power. The valve opening change step size constraint is used to limit the single adjustment range to avoid water hammer or excessively rapid pressure fluctuations in the pipeline network.

[0059] The aforementioned preset values, such as the time series length, valve opening change step size, safe pressure range, and pump station rated power settings, are all determined based on the pipeline design pressure and historical equipment operation data. At the same time, the pipeline's resistance to water hammer, flow regulation stability, and short-term water demand characteristics are also taken into account to ensure that the phased regulation plan meets water supply demand while ensuring pipeline safety and stable pump station operation, thus achieving executable and refined water supply control.

[0060] The current operating power of each pumping station is obtained by real-time acquisition of instantaneous power and start / stop status via on-site sensors, and the operating parameters of the pumping stations are recorded. The acquired power value is compared with the rated power of the pumping station to determine whether the start / stop and output power of each pumping station in the regulation plan exceed the upper limit of the rated power. The specific judgment method is as follows: the output power of each pumping station required by the regulation plan in each stage is compared with the rated power one by one. If the power required by the regulation plan exceeds the rated upper limit, the power of that pumping station is determined to be out of limit and it is marked as an object requiring adjustment; if the power does not exceed the rated value, it is determined to be executable.

[0061] Simultaneously, simulation calculations were performed on the network pressure corresponding to each water monitoring unit in the phased regulation plan. During the simulation, the start-up and shutdown sequence of each pump station and the valve opening adjustment range in the plan were input into a network simulation tool, such as EPANET software. A topology model consistent with the actual water supply network was constructed in the software, and pump station characteristic curves, pipeline parameters, and node water loads were set. Extended-cycle hydraulic calculations were performed according to the set time step to obtain the pressure values ​​and trends of each monitoring node at different stages. The calculation results were compared with the preset safe pressure range one by one to determine whether the regulation scheme met the safe operation requirements. The preset safe pressure range was set based on the network design pressure, historical operating experience, and the allowable pressure fluctuation of the equipment. For example, if the design pressure is 2.5 MPa and the allowable fluctuation is ±0.2 MPa, then the safe range is 2.3 to 2.7 MPa. If the simulation results show that the pressure of a certain node is lower than 2.3 MPa or higher than 2.7 MPa, then the pressure of that node is determined to be out of limit.

[0062] When pump station power or pipeline pressure exceeds limits, the corresponding pump station or valve adjustment item is marked as an over-limit item, and the operator is prompted to check or correct it. The operator can adjust the pump station output power, start-up / shutdown sequence, or valve opening based on the marking information, recalculating power and pressure until all over-limit items are corrected, forming a safe and executable phased adjustment plan. If no over-limit items are found, the plan is confirmed to be executable. The start-up / shutdown sequence, output power, and valve opening adjustment range of each pump station are generated into a final scheduling command, which is sent to the corresponding pump and valve controller via the NB-IoT downlink. The phased flow regulation and pressure control operations are executed according to the plan, achieving safe and stable operation of the local water supply.

[0063] Scenario 2: Predict short-term water demand within the current water monitoring period and execute a steady-state confirmation prediction process for water usage trends, specifically:

[0064] The existing time-series forecasting model is invoked and adaptively updated by combining real-time water usage data from each water monitoring unit and operational data from each monitoring element under the current water monitoring scenario. This results in a water demand forecasting model suitable for the current water monitoring scenario. During the adaptive update process, real-time collected monitoring data includes instantaneous flow rate, cumulative water volume, pipeline pressure, valve opening status, pump station operating power, and start / stop status for each water monitoring unit. By inputting this data into the existing time-series model and adjusting the model parameters, the forecast results accurately reflect the actual water usage characteristics within the current water monitoring cycle, enhancing the adaptability and accuracy of short-term water demand forecasting. The updated water demand forecasting model uses water monitoring data from each monitoring unit as input and outputs short-term water demand forecasts for the current water monitoring cycle through time-series calculations, providing a data foundation for water supply regulation in each monitoring unit.

[0065] For the short-term water demand forecasts over 10 consecutive water monitoring cycles, the values ​​are first sorted according to time series, then smoothed to eliminate occasional fluctuations and noise, resulting in a continuous and smooth water demand trend curve. Subsequently, the difference in forecasted water demand between adjacent monitoring cycles is calculated to form the forecasted water demand trend, thereby quantifying the magnitude and direction of increase or decrease in water supply demand between cycles. To further analyze the water demand trend, a linear fit is performed on the calculated forecasted water demand trend to obtain the trend slope, which reflects the speed and direction of short-term water demand changes.

[0066] Based on a preset range of trend slope values, the trend slope is mapped to a trend discrimination parameter. Defining the value range transforms the continuously quantified trend slope into a discrete state indicator, indicating the direction and state of water demand changes. When the trend discrimination parameter is greater than zero, it is 1, indicating an upward trend in current water demand, suggesting an increase in local water supply demand. When the trend discrimination parameter is equal to zero, it is 0, indicating a steady-state level in current water demand, meaning the demand change is not significant. When the trend discrimination parameter is less than zero, it is -1, indicating a downward trend in current water demand, reflecting a decrease in short-term water supply demand. Through the trend discrimination parameter, personnel can convert continuous short-term forecasts into clear upward, steady-state, or downward states, providing an intuitive decision-making basis for water supply scheduling and pump / valve control. This enables dynamic monitoring and refined management of short-term water demand, ensuring that local water supply capacity matches actual demand, while improving the response speed and accuracy of water supply regulation.

[0067] This invention provides an intelligent scheduling system for urban water supply networks based on an Internet of Things (IoT) cloud platform, such as... Figure 8 The diagram shown illustrates the structure of an intelligent dispatching system for urban water supply networks based on an IoT cloud platform. This system may include:

[0068] The monitoring unit construction and validity verification module is used to divide the urban water supply network into several water monitoring units. Through NB-IoT, the water usage data and equipment operation data of each water monitoring unit are connected to the One NET IoT cloud platform to construct the internal state variables of the water monitoring unit and verify its validity.

[0069] The local water supply determination module is used to determine whether the local water supply meets the predicted demand by analyzing the changes in flow and pressure of each water monitoring unit during each water monitoring cycle.

[0070] If it is determined that the local water supply does not meet the predicted demand, the system will proceed to the multi-step scheduling management module; if it is determined that the local water supply meets the predicted demand, the system will proceed to the steady-state trend confirmation module.

[0071] The multi-step scheduling management module is used to perform multi-step predictive calculations for water supply scheduling and to adjust valve opening and / or pump station start-up and shutdown sequence in stages through the downlink of NB-IoT.

[0072] The steady-state trend confirmation module is used to predict short-term water demand within the current water monitoring cycle and to execute a steady-state confirmation prediction process for water use trends.

[0073] In summary, this invention constructs a smart water management technology system covering the entire chain of data acquisition, network transmission, cloud platform processing, and business applications. It divides the urban water supply network into refined water monitoring units, collecting key variables such as instantaneous flow rate, cumulative water volume, network pressure, and pump / valve status in real time. These data are then connected to the One NET cloud platform for validity verification and instability / fluctuation determination, achieving accurate perception and proactive early warning of network operation status. Based on this, the system comprehensively evaluates the local water supply capacity and demand matching degree through integral weighted calculation of water supply determination parameters. For different scenarios, it executes multi-step predictive scheduling and short-term steady-state confirmation processes for water demand. In the multi-step predictive scheduling, a phased adjustment plan is generated based on trend function fitting and accurately executed via the NB-IoT downlink after verification by pump station power and network pressure constraints. In water demand prediction, an adaptively updated time series model is used to output predicted values ​​and perform trend slope mapping to quantify water usage patterns.

[0074] This method improves the accuracy of hourly water demand forecasting, effectively solves the problem of delayed dispatch response caused by forecast lag during the summer peak water consumption period, significantly reduces pipeline pressure fluctuations and frequent pump station start-ups and shutdowns through phased and refined regulation, reduces water supply energy consumption and pipeline leakage rate, and ensures stable water pressure for high-rise users. It achieves deep integration of forecast results and dispatch control, and provides technical support for the safe and efficient operation of smart water management in cities.

[0075] It should be added that, such as Figure 7 The diagram shows the architecture of the water demand prediction model. This model takes urban water demand time series as input and first decomposes the original time series using T-EMD (Temporal Empirical Mode Decomposition) to obtain multiple intrinsic mode functions (IMF1 to IMFQ). Each IMF represents the variation characteristics of different frequency components in the time series. The decomposed IMF sequences are then processed by ranking entropy calculation, and the prediction model parameters are fine-tuned using numerical optimization algorithms to form training parameters suitable for different mode components. Based on the ranking entropy threshold, the system selects different prediction methods: if the ranking entropy is less than the threshold, the corresponding IMF is predicted using a linear sequence model (LSM); if the ranking entropy is greater than the threshold, a long short-term memory network (LSTM) is used for prediction.

[0076] After each IMF (Information Frequency Component) is calculated using a prediction method, its corresponding short-term water demand forecast is obtained. Subsequently, the forecast results of all IMFs are merged, integrating the forecast information of each frequency component to form a final comprehensive forecast result, which is used to output the short-term water demand forecast value within the current water monitoring cycle. During model training, LSTM uses historical water demand time series and real-time operational data from each monitoring unit as input for iterative training, and adjusts network parameters through a gradient optimization algorithm to minimize prediction error; LSM obtains linear model parameters through least squares fitting. The overall architecture realizes multimodal decomposition of urban water demand time series, modal feature selection, targeted model prediction, and result fusion, ensuring accurate prediction of water demand data with different fluctuation characteristics, while taking into account both high-frequency fluctuations and low-frequency trends, providing a scientific basis for short-term water management and water supply regulation.

[0077] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can 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.

[0078] This invention is described with reference to flowchart illustrations and / or block diagrams of systems, apparatus (systems), and computer program products according to embodiments of the invention. 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 illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0079] 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.

[0080] 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.

[0081] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the invention.

[0082] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for intelligent scheduling of urban water supply networks based on an Internet of Things (IoT) cloud platform, characterized in that: The method includes: The urban water supply network is divided into several water monitoring units. The water usage data and equipment operation data of each water monitoring unit are connected to the One NET IoT cloud platform through NB-IoT to construct the internal state variables of the water monitoring units and verify their effectiveness. The internal state variables include at least instantaneous flow rate, cumulative water volume, pipeline pressure value, and pump and valve status; Within each water monitoring cycle, by analyzing the changes in flow and pressure of each water monitoring unit, it is determined whether the local water supply meets the predicted demand. If it is determined that the local water supply does not meet the predicted demand, multi-step predictive calculations for water supply scheduling are performed, and the valve opening and / or pump station start-up and shutdown sequence are adjusted in stages through the downlink of NB-IoT. If it is determined that the local water supply meets the predicted demand, then the short-term water demand within the current water monitoring cycle is predicted, and the steady-state confirmation prediction process for water use trends is executed. The determination of whether the local water supply meets the predicted demand includes: Within each water monitoring cycle, the flow and pressure changes of each water monitoring unit are monitored, and the area in the water supply network that needs to be determined and its corresponding monitoring unit set are recorded as the local water supply area. The instantaneous flow sequence within the local water supply area is accumulated and summed, and the result is recorded as the cumulative flow value. The time derivative of the pipeline pressure sequence within the local water supply area is processed, and the result is recorded as the pressure change rate. The cumulative flow rate and the pressure change rate are periodically integrated or cumulatively calculated to obtain water supply judgment parameters used to determine the degree of matching between local water supply capacity and preset demand. If the water supply determination parameters exceed the preset water supply determination parameters, it is determined that the local water supply does not meet the predicted water demand; otherwise, it is determined that the local water supply meets the predicted water demand. The predicted short-term water demand during the current water monitoring cycle includes: The stored time series prediction model is invoked, and combined with the real-time water consumption data and the operation data of each monitoring element under the current water monitoring scenario, the time series prediction model is adaptively updated to obtain a water demand prediction model suitable for the current water monitoring scenario. Water monitoring data from each water monitoring unit is used as input to the water demand prediction model. Time series calculations are performed using the water demand prediction model to output short-term water demand prediction values ​​for the current water monitoring cycle. The water monitoring data includes at least the instantaneous flow rate, cumulative water volume, pipeline pressure, valve opening status, pump station operating power, and start / stop status of each water monitoring unit. The steady-state confirmation and prediction process for water usage trends includes: The short-term water demand forecast values ​​within a consecutive preset number of water monitoring cycles are sorted and smoothed in time sequence to obtain the water demand change trend curve, and the predicted water demand change between adjacent water monitoring cycles is calculated to obtain the predicted water demand change trend. A linear fitting analysis is performed on the predicted water demand change trend to calculate the corresponding trend slope, and the predicted water demand change trend is mapped to a trend discrimination parameter according to the defined range of the trend slope. The defined value range indicates the direction and state of water demand change corresponding to the trend slope; The steady-state confirmation and prediction process for water usage trends also includes: When the trend discrimination parameter is greater than zero, the trend discrimination parameter takes the value of 1, indicating that the current water demand is showing an upward trend; When the trend discrimination parameter is equal to zero, the trend discrimination parameter takes a value of 0, indicating that the current water demand is at a steady state level; When the trend discrimination parameter is less than zero, the trend discrimination parameter takes the value of -1, indicating that the current water demand is showing a downward trend.

2. The intelligent scheduling method for urban water supply networks based on an IoT cloud platform as described in claim 1, characterized in that, The construction of the internal state variables of the water monitoring unit and the verification of its effectiveness include: Based on the basic water supply attribute information preset in the urban water supply network, state parameters are constructed for each monitoring element in each water monitoring unit to form corresponding internal state variables. The monitoring elements include at least flow meters, pressure transmitters and pump valve controllers. Under the constraint of a preset flow rate change step size, a metering consistency constraint is established between the integral results of cumulative water volume and instantaneous flow rate within the same metering period. The preset flow rate change step size represents the maximum allowable flow rate change amplitude per unit time. Based on the aforementioned metering consistency constraints, and combined with the pressure characteristics of the pipeline nodes corresponding to the preset basic water supply attribute information, the pressure fluctuation range, pressure reference threshold, and pressure regulation response time limit of the pipeline pressure value are determined to characterize the stability of the pipeline pressure state. Determine whether the stability level exhibits a first or second unstable fluctuation. The first unstable fluctuation indicates an operating state where the stability of the pipeline pressure is higher than the upper limit of the preset stability range, and the second unstable fluctuation indicates an operating state where the stability of the pipeline pressure is lower than the lower limit of the preset stability range.

3. The intelligent scheduling method for urban water supply networks based on an IoT cloud platform as described in claim 2, characterized in that, The process of constructing the internal state variables of the water monitoring unit and verifying their effectiveness also includes: If there is no first or second unstable fluctuation, continue to monitor the pipeline pressure status; When the first unstable fluctuation occurs, the corresponding first management strategy is triggered. By issuing a flow suppression and adjustment command, the pump and valve controller is controlled to reduce the valve opening and / or reduce the output power of the pump station, so that the water supply flow rate decreases step by step according to the preset flow rate change step. When a second unstable fluctuation occurs, the corresponding second management strategy is triggered. By issuing a pressure compensation adjustment command, the pump and valve controller is controlled to increase the valve opening and / or increase the output power of the pump station, so that the water supply flow rate increases step by step according to the preset flow rate change step. After completing the adjustment operations corresponding to the first or second management strategy, if there is no first or second unstable fluctuation, the pipeline pressure status will continue to be monitored; otherwise, an early warning of unstable fluctuation risk will be issued.

4. The intelligent scheduling method for urban water supply networks based on an IoT cloud platform as described in claim 1, characterized in that, The specific steps for performing multi-step prediction calculations for water supply scheduling are as follows: The instantaneous flow rate, cumulative water volume, and pipeline pressure values ​​of the local water supply area are sorted and fitted according to the preset time series to determine the trend function of each water monitoring unit under the current operating state. The trend function represents the variation pattern and rate of change of flow and pressure of each water monitoring unit over time in the current period, and is used to reflect the operating trend of the short-term water supply process; Based on the trend function, calculate the water supply and pressure adjustment required by each water monitoring unit to meet short-term water supply demand in the current cycle, and generate the corresponding valve opening adjustment range and pump station start-up and shutdown sequence. The phased adjustment plan formed by the valve opening adjustment range and the pump station start-up and shutdown sequence is constrained and verified, including the pump station operating power limit and / or valve opening change step size constraint.

5. The intelligent scheduling method for urban water supply networks based on an IoT cloud platform as described in claim 4, characterized in that, The specific steps for adjusting valve opening and / or pump station start-up and shutdown sequence in stages via the NB-IoT downlink are as follows: Obtain the current operating power of each pumping station and combine it with its rated power to determine whether the start-up, shutdown and output power of each pumping station in the phased adjustment plan exceed the upper limit of the rated power. Simulation calculations are performed on the pipeline pressure corresponding to each water monitoring unit in the phased regulation plan to determine whether the pipeline pressure is within the preset safe range. If the current operating power exceeds the rated power limit and / or the pipeline pressure is not within the preset safety range, the corresponding pump station and / or valve adjustment item will be recorded as an over-limit item, and the preset personnel will be prompted to inspect it. After the over-limit item is corrected, a new phased adjustment plan will be formed. Conversely, if the constraints are not met, a scheduling command that has been verified by constraints is output and sent to the corresponding pump and valve controller via the NB-IoT downlink to execute the corresponding phased flow regulation and pressure control operations.

6. An intelligent scheduling system for urban water supply networks based on an Internet of Things (IoT) cloud platform, employing the intelligent scheduling method for urban water supply networks based on an IoT cloud platform as described in any one of claims 1-5, characterized in that... include: The monitoring unit construction and validity verification module is used to divide the urban water supply network into several water monitoring units. Through NB-IoT, the water consumption data and equipment operation data of each water monitoring unit are connected to the One NET IoT cloud platform to construct the internal state variables of the water monitoring unit and verify its validity. The local water supply determination module is used to determine whether the local water supply meets the predicted demand by analyzing the changes in flow and pressure of each water monitoring unit during each water monitoring cycle. If it is determined that the local water supply does not meet the predicted demand, the process will proceed to the multi-step scheduling management module; if it is determined that the local water supply meets the predicted demand, the process will proceed to the steady-state trend confirmation module. The multi-step scheduling management module is used to perform multi-step predictive calculations for water supply scheduling, and to adjust valve opening and / or pump station start-up and shutdown sequence in stages through the downlink of NB-IoT; The steady-state trend confirmation module is used to predict the short-term water demand within the current water monitoring cycle and to execute the steady-state confirmation prediction process for water use trends.

Images