Time-phased intelligent pressure regulation management method and system for secondary water supply

Abstract

The invention discloses a time-phased intelligent pressure regulation management method and system for secondary water supply, and belongs to the technical field of water supply management, and the method specifically comprises the steps: dividing a day into a water consumption peak period and a water consumption low period based on historical water consumption flow data, and determining the starting and ending time nodes of each time period; through hydraulic calculation and actual water supply pressure monitoring, a pressure control point at the tail end of the water supply system is locked, and water supply pressure data of the point are collected in real time; in the water consumption peak period, a peak pressurization control strategy is executed; in the water consumption low-peak period, analysis is conducted according to historical water supply pressure data of a tail end pressure control point, and a low-peak period pressure safety threshold value is set; a pressure instruction is sent to a control system according to the threshold value, and water supply equipment is controlled to output pressure according to the threshold value; the water supply pressure of a tail end pressure control point is monitored in real time, and when the pressure is lower than a low-peak-period pressure safety threshold value, pre-alarm is triggered, and pressure changes are continuously recorded; and if the pre-alarm state is not finished continuously, switching to a peak-period peak pressurization control strategy.

Description

Technical Field

[0001] The present invention relates to the technical field of water supply management, and in particular to a time-divided intelligent pressure regulation management method and system for secondary water supply. Background Art

[0002] In the water supply system of urban buildings and residential areas, the secondary water supply system plays the key role of pressurizing and delivering municipal water to the user end. However, traditional secondary water supply systems have obvious limitations in pressure control: to ensure stable water supply during peak water usage periods, the system often adopts a fixed higher pressure setting. This "one-size-fits-all" model will cause the water supply pressure to far exceed the actual demand during non-peak hours, which not only causes significant waste of water pump energy consumption, but may also aggravate pipeline losses due to long-term high-pressure operation and increase maintenance costs. At the same time, some systems lack real-time monitoring of the pressure at the terminal pressure control point, making it difficult to accurately perceive changes in the terminal water pressure, and are prone to unstable or insufficient pressure, affecting the user's water comfort.

[0003] Energy-saving retrofits for traditional secondary water supply systems face numerous challenges. First, large-scale equipment replacement is not only costly but can also disrupt normal water supply. Second, existing control strategies rely heavily on manual experience, making it difficult to dynamically adjust pressure based on water usage patterns, thus failing to balance water supply safety with energy conservation. Against this backdrop, the key to addressing the high energy consumption and low efficiency of secondary water supply systems is finding intelligent solutions that achieve precise time-based pressure regulation without adding expensive equipment. This approach also provides a new technical path for energy-efficient upgrades in urban water supply systems. Summary of the Invention

[0004] The purpose of the present invention is to provide a time-divided intelligent pressure regulation management method and system for secondary water supply, which solves the following technical problems: Large-scale replacement of equipment is not only costly, but may also affect normal water supply; on the other hand, existing control strategies rely heavily on manual experience, making it difficult to dynamically adjust pressure according to water usage patterns, and unable to balance water supply safety and energy-saving goals.

[0005] The purpose of the present invention can be achieved through the following technical solutions: In a first aspect, a time-divided intelligent pressure regulation management method for secondary water supply comprises the following steps: S1. Based on historical water flow data, divide the day into peak water use period and low water use period, and determine the start and end time nodes of each period; S2. Lock the pressure control point at the end of the water supply system through hydraulic calculation and actual water supply pressure monitoring, and collect water supply pressure data at this point in real time. The process of locking the pressure control point at the end of the water supply system is as follows: S21. Build a digital twin model of the water supply system; S22. Based on the digital twin model, input historical typical operating conditions covering water flow data in different time periods and seasons, conduct steady-state and transient hydraulic simulations using hydraulic calculation software, output the theoretical pressure value of each node, and generate a multi-operating condition theoretical pressure field distribution. Mark the candidate node set with the lowest theoretical pressure under each operating condition. S23. Based on the distribution characteristics of the candidate nodes in the theoretical pressure field, monitoring points are arranged along the pipeline and at the user end; S24. During typical water usage periods, covering both peak and off-peak periods, the digital twin model and the dynamic monitoring network are run simultaneously to collect theoretical and actual pressure data in the same time series. Error analysis is performed by calculating the deviation between the theoretical and actual values. The pipe resistance coefficient and node flow distribution parameters in the model are adjusted until the deviation between the pressure distribution output by the model and the actual monitoring results is within a preset range. S25. Based on the calibrated model and combined with real-time pressure data from the dynamic monitoring network, extract the lowest pressure value of each node throughout the entire operating cycle, the duration of pressure below the preset basic threshold, and the pressure fluctuation amplitude. A comprehensive score is then assigned to all nodes using a multi-feature weighted algorithm. The algorithm weights are set based on the water supply safety priority, and the node with the lowest score is designated as the terminal pressure control point of the water supply system. S3. During the peak water consumption period, the peak pressure boost control strategy is implemented; S4. During the low-peak water consumption period, the historical water supply pressure data of the terminal pressure control point is analyzed to set a low-peak pressure safety threshold; based on the threshold, a pressure instruction is sent to the control system to control the output pressure of the water supply equipment according to the threshold; S5. Real-time monitoring of the water supply pressure at the terminal pressure control point. When the pressure is lower than the low-peak pressure safety threshold, a pre-alarm is triggered and pressure changes are continuously recorded. If the pre-alarm state persists, the control system switches to the peak-peak boost control strategy. After the water supply pressure at the terminal pressure control point returns to normal, the low-peak pressure safety threshold for that period is automatically corrected.

[0006] As a further solution of the present invention: in step S1, based on the historical water flow data, a day is divided into a peak water consumption period and a low water consumption period, and the process of determining the start and end time nodes of each period is as follows: S11. Collect 30 consecutive days of raw water flow data, convert it into a continuous time series with a time unit of 15 minutes, remove extreme outliers through the box plot method, and use linear interpolation to fill in the missing data segments; S12. Calculate the flow rate value for each time unit, calculate the flow rate average within the window based on a 2-hour sliding window, average the 30-day data at the same time unit, and generate a daily average flow rate distribution curve; S13. Calculate the overall mean and standard deviation based on the daily average traffic distribution curve, delineate a candidate high-peak interval that is higher than the mean by +1.2 times the standard deviation, and a candidate low-peak interval that is lower than the mean by -1.2 times the standard deviation. Iteratively adjust the standard deviation multiple to ensure that the candidate intervals cover more than 90% of the significant fluctuation segments of the curve. S14. Traverse the daily average traffic distribution curve, identify consecutive peak and low-peak candidate time unit clusters and mark them as preliminary time periods, and merge preliminary time periods of the same type with an interval of no more than 1 time unit; S15. Count the frequency of occurrence of each preliminary time period that meets the corresponding traffic characteristics in multi-day data, retain stable time periods with a frequency of not less than 75%, and use the time when three consecutive time units meet the characteristics as the start and end nodes to determine the time range of peak and off-peak periods.

[0007] As a further solution of the present invention: in step S23, the monitoring points are arranged so that the density of the candidate node set area is higher than that of other areas, and all monitoring points have real-time data transmission capabilities to collect actual operating pressure data of each node.

[0008] As a further solution of the present invention: in step S3, during the water consumption peak period, the specific manner of executing the peak boost control strategy is as follows: A fixed peak water supply pressure threshold is preset, and real-time pressure data of the water supply system is obtained through pressure monitoring; the real-time pressure data is compared with the peak water supply pressure threshold. If the real-time pressure is lower than the threshold, the output pressure is increased; if the real-time pressure is higher than the threshold, the output pressure is reduced; according to the comparison results of the real-time pressure data and the peak water supply pressure threshold, corresponding adjustments are made to ensure that the water supply system pressure is always maintained within the peak water supply pressure threshold range.

[0009] As a further solution of the present invention: in step S4, during the low-peak period of water use, the historical water supply pressure data of the terminal pressure control point is analyzed to set the low-peak period pressure safety threshold as follows: S41. Collect the low-peak water supply pressure data of the terminal pressure control points for the past 30 days and classify them by date to form a historical water supply pressure data set of the terminal pressure control points; S42. Based on the determined start and end times of the off-peak period and in combination with historical water flow fluctuation characteristics, the off-peak period is further divided into several off-peak sub-periods; each off-peak sub-period has a fixed duration, and the water flow fluctuation amplitude within each sub-period is within a preset range; S43. For each off-peak period, extract the pressure characteristic parameters of the corresponding period from the historical water supply pressure data set of the terminal pressure control point, including the pressure mean value, pressure minimum value, pressure standard deviation, and continuous stable pressure duration within the sub-period; S44. For each off-peak period, the pressure baseline value is calculated using the weighted average method: S45. Analyze the difference distribution between the historical minimum pressure value and the pressure reference value of each off-peak period, and set a corresponding safety margin factor for each off-peak period in combination with the minimum service head requirement of the water supply system; S46. For each off-peak period, multiply its pressure baseline value by the safety margin coefficient to obtain the off-peak pressure safety threshold of the sub-period; summarize the thresholds of all off-peak sub-periods to form a complete off-peak pressure safety threshold system.

[0010] As a further solution of the present invention: in step S44, it also includes: The pressure average value of the sub-period is weighted by the continuous stable pressure duration in the sub-period; the weight distribution rule is: the pressure average value with a continuous stable duration longer than the preset duration is given a higher weight.

[0011] As a further solution of the present invention: in step S5, the process of automatically correcting the low-peak pressure safety threshold of the period is: S51: When the water supply pressure at the terminal pressure control point rises back above the original low-peak pressure safety threshold and the continuous stability time reaches the preset pressure stability time, it is determined that the pressure has returned to normal; S52. Extract the complete data sequence of this pre-alarm event, including: the duration of the pressure below the threshold, the pressure fluctuation amplitude, the rate of pressure recovery after switching to the peak period strategy, the pressure average value in the first complete off-peak period after returning to normal, and the pressure stable duration; S53. Calculate the difference between the mean pressure value of the first complete off-peak period after normal operation and the original off-peak period pressure safety threshold to obtain the actual pressure deviation value. Combined with the maximum magnitude of the pressure below the threshold during this abnormal event, generate a deviation coefficient using a proportional algorithm. S54. Determine a dynamic correction factor based on the deviation coefficient and adjust the safety margin coefficient of the original off-peak period: if the actual pressure deviation value is positive, adjust the safety margin coefficient downward; if it is negative, adjust the safety margin coefficient upward. The adjustment range is determined by the dynamic correction factor. S55. Recalculate the adjusted safety margin coefficient and the original pressure reference value to obtain a revised pressure safety threshold for the off-peak period. The revised threshold satisfies the constraint condition of being no less than the minimum service head of the water supply system.

[0012] In a second aspect, a time-divided intelligent pressure regulation management system for secondary water supply is provided, which is used to implement the above-mentioned time-divided intelligent pressure regulation management method for secondary water supply, comprising: The water usage period division module divides a day into peak water usage period and low water usage period based on historical water flow data, and determines the start and end time nodes of each period; The terminal pressure control point locking module locks the terminal pressure control point of the water supply system through hydraulic calculation and actual water supply pressure monitoring, and collects the water supply pressure data of the point in real time; Peak boost control module, used to execute peak boost control strategy during peak water consumption period; The low-peak pressure threshold setting module is used to analyze the historical water supply pressure data of the terminal pressure control point during the low-peak water consumption period and set the low-peak pressure safety threshold; based on the threshold, it sends a pressure instruction to the control system to control the output pressure of the water supply equipment according to the threshold; The terminal pressure monitoring and correction module is used to monitor the water supply pressure of the terminal pressure control point in real time. When the pressure is lower than the low-peak pressure safety threshold, it triggers a pre-alarm and continuously records the pressure changes. If the pre-alarm state continues, the control system switches to the peak-peak boost control strategy. After the water supply pressure of the terminal pressure control point returns to normal, the low-peak pressure safety threshold of the period is automatically corrected.

[0013] Beneficial effects of the present invention: This invention can effectively avoid the high costs and water supply interruption risks associated with large-scale equipment replacement, achieving energy-saving transformation through strategic optimization of existing water supply systems. Its core lies in dividing peak and off-peak periods based on historical water usage data, and formulating targeted pressure regulation strategies in combination with real-time pressure monitoring of terminal pressure control points. No expensive new equipment is required throughout the process. Through data analysis and optimization of intelligent control logic, the system can achieve precise pressure regulation in different time periods based on the existing hardware. This transformation model not only reduces initial investment but also avoids water supply fluctuations that may occur during equipment replacement, ensuring that energy-saving upgrades and normal water supply do not interfere with each other.

[0014] At the same time, the present invention breaks away from the reliance on manual experience and balances water supply safety and energy conservation through dynamic intelligent regulation. On the one hand, the pressure boosting strategy is automatically executed during peak periods to ensure stable water supply during periods of concentrated water use; on the other hand, during off-peak periods, a safety threshold is set based on the historical data of the terminal pressure control point, and the pressure output is reduced according to actual demand to reduce energy consumption. More importantly, the real-time monitoring and automatic correction mechanism can quickly switch to the guarantee mode when the pressure is abnormal, and dynamically optimize the threshold after returning to normal, so that the system always adapts to changes in water use patterns, avoiding the lag and errors of manual regulation and achieving a balance between safety priority and energy efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The present invention will be further described below with reference to the accompanying drawings.

[0016] Figure 1 The present invention is a flowchart of a time-divided intelligent pressure regulation management method for secondary water supply according to an embodiment of the present invention.

[0017] Figure 2 The present invention is a schematic structural diagram of a time-divided intelligent pressure regulation management system for secondary water supply according to an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0019] Example 1 See also Figure 1 As shown, the present invention is a time-divided intelligent pressure regulation management method for secondary water supply, comprising: S1. Based on historical water flow data, divide the day into peak water use period and low water use period, and determine the start and end time nodes of each period.

[0020] In step S1 of this embodiment, based on historical water flow data, a day is divided into a peak water consumption period and a low water consumption period. The process of determining the start and end time nodes of each period is as follows: Collect 30 consecutive days of raw water flow data, and continuously record the water flow information at every moment during these 30 days; convert these raw data into a continuous time series with 15 minutes as the time unit. Specifically, divide the 24 hours a day into 96 15-minute time periods, and then integrate all the flow data in each time period into the flow value corresponding to the time period, forming a continuous data series arranged in chronological order. This is done to unify the time granularity of the data, so as to facilitate the subsequent comparison and analysis of water use patterns on different dates. Extreme outliers are eliminated through the box plot method. The box plot will first determine the quartiles of the data. The interval between the upper quartile and the lower quartile contains most of the normal data. Values ​​that exceed a certain range of this interval are considered extreme outliers. These values ​​may be caused by monitoring equipment failures, sudden non-water use factors, etc., and are not real water use conditions. Eliminating them can avoid these abnormal data from interfering with the overall analysis results; linear interpolation is used to fill in the missing data segments. The flow rate value for each time unit is calculated. This involves statistically processing the raw flow data within each 15-minute time unit to obtain a representative flow value for that time period, thereby reflecting water consumption within that 15-minute period. The average flow rate within a 2-hour sliding window is calculated. A 2-hour window consists of eight 15-minute time units. A sliding window starts from the first time unit and moves backward one 15-minute time unit at a time to form a new window. The flow rate values ​​of the eight time units within each window are averaged. This smooths out short-term flow fluctuations and more clearly demonstrates water consumption trends over longer time periods. The data for 30 days are averaged within the same time unit. That is, for the same 15-minute time unit in each day (for example, 00:00-00:15 every day), the average flow rate for that time unit is calculated over these 30 days. This method eliminates the impact of unique daily water consumption and produces more general flow data. These average values ​​calculated by time unit are arranged in chronological order to generate a daily average flow distribution curve, which can intuitively show the changes in average water consumption flow at different time periods throughout the day.

[0021] The overall mean and standard deviation are calculated based on the daily average flow distribution curve. The overall mean is the average of the flow averages of all 15-minute time units in the curve, representing the overall average water flow level for the day. The standard deviation is used to measure the degree to which these flow averages deviate from the overall mean, reflecting the fluctuation of water flow during the day. Peak candidate intervals are defined as being higher than the mean + 1.2 times the standard deviation, and low-peak candidate intervals are defined as being lower than the mean - 1.2 times the standard deviation. This is because when the flow value is higher than the overall mean plus 1.2 times the standard deviation, it indicates that the flow in that time period is significantly higher than the average level, and is likely to be a peak period with more concentrated water use. When the flow value is lower than the overall mean minus 1.2 times the standard deviation, it indicates that the flow in that time period is significantly lower than the average level, and is likely to be a low-peak period with less water use. By iteratively adjusting the standard deviation multiple, the candidate interval covers more than 90% of the significant fluctuation segments of the curve. Since the initially set 1.2 times the standard deviation may not cover most of the obvious flow fluctuation parts in the curve, this multiple needs to be continuously adjusted until the peak and low peak candidate intervals can include more than 90% of the fluctuation parts of the curve that are significantly deviated from the average level. This is done to ensure that the candidate interval can effectively capture the main water flow change periods in a day and provide a reliable basis for subsequent time period division.

[0022] The daily average flow distribution curve is traversed, and the flow values ​​for each 15-minute time unit on the curve are examined in chronological order to determine whether they fall within the previously defined peak or low-peak candidate intervals. Consecutive clusters of peak and low-peak candidate time units are identified and marked as preliminary time periods. Specifically, when multiple consecutive 15-minute time units are found to fall within the peak candidate interval, these consecutive units are grouped together and marked as a preliminary peak period. Similarly, clusters of time units that fall within the low-peak candidate interval are marked as preliminary low-peak periods, as these clusters better reflect a coherent peak or low-peak state. Preliminary time periods of the same type separated by no more than one time unit are merged. For example, if two preliminary peak periods are separated by only one 15-minute time unit (which does not fall within the peak candidate interval), this gap may be caused by a brief fluctuation in water use. In reality, the two periods still belong to the same overall peak cycle. Merging them more accurately reflects the continuous peak or low-peak periods, avoiding the fragmentation of a complete period due to a small gap.

[0023] The frequency of occurrence of each preliminary time period that meets the corresponding traffic characteristics in the multi-day data was counted. Specifically, for each preliminarily defined peak or off-peak period, the number of days in the 30-day raw data corresponding to that period's time units actually met the peak or off-peak traffic characteristics (i.e., fell within the respective candidate intervals) was examined. The frequency of occurrence was calculated as the ratio of the number of days meeting the characteristics to the 30-day period. Stable time periods with a frequency of at least 75% were retained, as such periods exhibit peak or off-peak characteristics on most days, indicating strong stability and regularity. Time periods with lower frequencies were excluded, likely occasional exceptions and unrepresentative. The time ranges of peak and off-peak periods were determined using the times when three consecutive time units met the characteristics as the start and end nodes. Since three consecutive 15-minute time units (i.e., 45 minutes) met the peak or off-peak characteristics, this indicated that the beginning or end of the period was stable and not caused by short-term traffic fluctuations. Using these start and end nodes allowed for more accurate definition of the specific time ranges of peak and off-peak periods, ensuring the reliability of the delineation results.

[0024] S2. Through hydraulic calculation and actual water supply pressure monitoring, the pressure control point at the end of the water supply system is locked, and the water supply pressure data at this point is collected in real time.

[0025] In step S2 of this embodiment, the process of locking the pressure control point at the end of the water supply system through hydraulic calculation and actual water supply pressure monitoring is as follows: Construct a digital twin model of the water supply system. Based on the actual physical structure of the water supply system, including detailed information such as the direction, diameter, and length of the pipelines, the location and parameters of the water pumps, the distribution of valves, and the location and distribution of users, a virtual model is created that is highly consistent with the actual system in structure and operating characteristics. This model can reflect the movement patterns of water flow and pressure changes in the actual system, providing an accurate virtual carrier for subsequent simulation analysis.

[0026] Based on the digital twin model, historical typical operating conditions covering water flow data in different time periods and seasons are input. The different time periods here include peak, flat and valley periods of the day. Different seasons take into account the impact of seasonal changes on water consumption. For example, water consumption in summer is usually higher than in winter. These representative historical operating condition data are selected because they can reflect the typical operating status of the system under different water use conditions; steady-state and transient hydraulic simulations are carried out through hydraulic calculation software. Steady-state simulation simulates the state of the system when the flow and pressure remain stable, which is suitable for analyzing relatively stable water use periods. Transient simulation simulates the process of dynamic changes in flow and pressure over time, which is suitable for analyzing water use. During rapidly changing periods, the combination of the two simulations can fully reflect the hydraulic characteristics of the system under various conditions; output the theoretical pressure value of each node and generate a multi-condition theoretical pressure field distribution. The hydraulic calculation software calculates the pressure loss of water when flowing in the pipeline based on the principles of conservation of energy and momentum in fluid mechanics, thereby obtaining the theoretical pressure value of each node. Presenting these pressure values ​​according to spatial positions forms a pressure field distribution, which can intuitively show the distribution of pressure in the system; mark the set of candidate nodes with the lowest theoretical pressure under each working condition. Under specific working conditions, the nodes with the lowest pressure are most likely to have insufficient water supply pressure. These nodes are potential terminal pressure control points and require special attention.

[0027] Based on the distribution characteristics of candidate nodes in the theoretical pressure field, monitoring points are deployed along the pipeline and at the user end. The distribution of candidate nodes in the theoretical pressure field can indicate which areas or locations are likely to have low pressure levels. Deploying monitoring points along the pipeline and at the user end in these areas is to obtain actual pressure data at these potential low-pressure points. Because theoretical simulation results may deviate from actual conditions, actual monitoring data can verify and calibrate the theoretical model, ensuring that subsequent analysis is based on real-world pressure conditions. The monitoring point layout ensures that the density of candidate node collection areas is higher than that in other areas, and all monitoring points have real-time data transmission capabilities to collect actual operating pressure data at each node.

[0028] The digital twin model and dynamic monitoring network were run simultaneously during typical water usage periods, encompassing both peak and off-peak periods. These two periods were chosen because they represent extreme states of system operation and can comprehensively test the model's accuracy. This synchronous operation ensures comparability between the pressure data calculated by the theoretical model and the pressure data measured at the monitoring points at the same time and under the same water usage conditions. After collecting theoretical and actual pressure data over the same time series, error analysis was performed by calculating the deviation between the theoretical and actual values. The magnitude of this deviation reflects the degree of agreement between the theoretical model and the actual system. The pipeline resistance coefficient and node flow distribution parameters in the model were adjusted. The pipeline resistance coefficient affects the pressure loss of water flowing through the pipeline; a larger resistance coefficient results in greater pressure loss. The node flow distribution parameters affect the flow rate at each node, resulting in different node pressures. These parameters may differ from their initial settings due to factors such as pipeline aging and changes in user water usage habits. Adjusting them can bring the theoretical model output closer to reality. The deviation between the model's output pressure distribution and the actual monitoring results falls within a preset range, indicating that the model accurately reflects the pressure conditions of the actual system and can be used for subsequent analysis.

[0029] Based on the calibrated model and combined with the real-time pressure data of the dynamic monitoring network, the calibrated model has been verified to reliably reflect the actual system status. Combined with the real-time monitoring data, the timeliness and accuracy of the information can be ensured. The lowest pressure value of each node in the entire operating cycle, the duration of the pressure below the preset basic threshold, and the pressure fluctuation amplitude are extracted. The entire operating cycle covers various water use conditions that the system may encounter. The lowest pressure value reflects the lowest pressure level that the node may reach. The lower the value, the more unfavorable the water supply; the longer the duration of the pressure below the preset basic threshold, the longer the node is in a state of insufficient water supply; the larger the pressure fluctuation amplitude, the more unstable the node water supply pressure. These characteristics are analyzed from different angles. It reflects the degree of water supply disadvantage of the node; all nodes are comprehensively scored through a multi-feature weighted algorithm. A single feature cannot fully evaluate the degree of disadvantage of the node. The weighted algorithm can comprehensively consider the influence of each feature; the algorithm weight is set according to the water supply safety priority. For example, the safety priority of the residential water node is higher than that of the industrial water node. Therefore, when scoring, the relevant features of the residential node will be given a higher weight to reflect its importance in water supply safety; the node with the lowest score is the terminal pressure control point of the water supply system, because this node shows the most serious water supply disadvantage after comprehensively considering factors such as the minimum pressure, insufficient time, and fluctuation amplitude. It is the node that needs the most attention and protection in the entire system.

[0030] S3. During the peak water usage period, the peak boost control strategy is implemented.

[0031] In step S3, during the peak water usage period, the specific method of executing the peak boost control strategy is as follows: A fixed peak water supply pressure threshold is preset. This threshold is determined based on the pressure demand at the most unfavorable point in the water supply system during peak hours and the safe operating pressure range of the pipe network. This ensures that all water points receive sufficient water pressure for normal use during peak water consumption periods (such as morning washing and evening cooking), while also preventing excessive pressure from causing pipe leaks or energy waste. Real-time collection of current pressure data from the water supply system is performed. These monitoring points can directly reflect the pressure status within the pipe network, ensuring that the data obtained is representative. Real-time pressure data is continuously compared against a preset peak water supply pressure threshold. This comparison process is key to determining whether the system's current pressure meets demand. If the real-time pressure falls below the threshold, it indicates insufficient water pressure within the network, potentially leading to insufficient water flow or even water supply disruption at remote water points. In this case, the control system must increase the output power of water supply equipment (such as pumps) or adjust the opening of pressure-regulating valves in the network to increase water flow pressure within the network and bring the real-time pressure closer to the threshold. If the real-time pressure exceeds the threshold, the current water supply pressure exceeds demand. Excessive pressure not only increases pump energy consumption but also may exacerbate pipe wear and aging. In this case, the output power of water supply equipment must be reduced or valves must be closed to reduce network pressure. This "monitoring-comparison-adjustment" closed-loop control approach dynamically responds to changes in peak water loads and promptly corrects pressure deviations, ensuring that the water supply system pressure remains within the preset peak water supply pressure threshold, ensuring stable and reliable peak water use.

[0032] S4. During the low-peak period of water use, analyze the historical water supply pressure data of the terminal pressure control point and set the low-peak pressure safety threshold; send a pressure instruction to the control system based on the threshold, and control the output pressure of the water supply equipment according to the threshold.

[0033] In step S4 of this embodiment, during the low-peak period of water use, the process of analyzing the historical water supply pressure data of the terminal pressure control point and setting the low-peak pressure safety threshold is as follows: Collect the low-peak water supply pressure data of the terminal pressure control point for the past 30 days. This requires the use of pressure monitoring equipment previously deployed at the terminal pressure control point to continuously record the real-time water supply pressure information during the low-peak period within these 30 days; classification by date is to classify the low-peak pressure data recorded every day into a separate category in order to distinguish the pressure conditions on different dates; forming a historical water supply pressure data set of the terminal pressure control point is to integrate these pressure data classified by date to form a complete data set containing 30 days of low-peak pressure information. The data volume of 30 days can cover a certain time range, including various water use conditions that may occur on different dates, providing sufficient samples for subsequent analysis, making the analysis results more representative and reliable.

[0034] Based on the determined start and end times of the low-peak period, that is, the start and end times of the low-peak period specified in the previous step, combined with the historical water flow fluctuation characteristics, that is, referring to the changes in water flow during the low-peak period recorded in the past, such as which time periods have large flow changes and which time periods have relatively stable flow, the low-peak period is further divided into several low-peak period sub-periods; the duration of each low-peak period is fixed, for example, it is set to 1 hour or 30 minutes, which facilitates unified analysis and comparison of different sub-periods; and the fluctuation amplitude of water flow in each sub-period is within the preset range. The fluctuation of water flow will directly affect the change of water supply pressure. Small flow fluctuation in the same sub-period means that the pressure change in this period is also relatively stable. Dividing such a time period into a sub-period can make the pressure characteristics of each sub-period more consistent, thereby laying the foundation for the subsequent setting of targeted pressure thresholds.

[0035] For each off-peak sub-period, the pressure characteristic parameters of the corresponding period are extracted from the historical water supply pressure data set of the terminal pressure control point. That is, from the integrated 30-day data, the pressure records of each sub-period on different dates are found, and then the relevant parameters of the sub-period are calculated, including the mean pressure in the sub-period, that is, the average value of all pressure data in the sub-period, reflecting the typical pressure level of the period; the minimum pressure, that is, the lowest pressure value in the sub-period, reflecting the lowest pressure condition that may occur in the period; the pressure standard deviation, used to measure the degree to which the pressure data in the sub-period deviates from the mean, reflecting the magnitude of pressure fluctuations; the continuous stable pressure duration, that is, the duration that the pressure in the sub-period remains within a certain stable range without obvious fluctuations, reflecting the stability of the pressure; these parameters are extracted because they describe the pressure characteristics of the sub-period from different dimensions, can comprehensively reflect the pressure condition of the period, and provide multi-faceted basis for the subsequent calculation of benchmark values.

[0036] For each off-peak sub-period, the weighted average method is used to calculate the pressure baseline value: the continuous stable pressure duration in the sub-period is used as the weight, and the pressure mean of the sub-period is weighted. That is, the pressure mean of the sub-period on different days is weighted according to their corresponding continuous stable duration. The pressure mean on the day with a long continuous stable duration accounts for a larger proportion in the calculation; the weight distribution rule is: the pressure mean with a continuous stable duration longer than the preset duration is given a higher weight. A long continuous stable pressure duration indicates that the pressure state of the sub-period on that date is more stable, better represents the normal pressure level of the sub-period, and is less affected by accidental fluctuations. Assigning a higher weight can make the calculated pressure baseline value closer to the actual and stable pressure level of the sub-period, thereby improving the reliability of the baseline value.

[0037] Analyze the distribution of the difference between the historical minimum pressure value and the pressure benchmark value of each low-peak sub-period, that is, calculate the difference between the minimum pressure value of each sub-period on different dates and the pressure benchmark value of the sub-period, and then observe the distribution of these differences to understand how much lower the minimum pressure value is usually than the benchmark value; combined with the minimum service head requirement of the water supply system, that is, the minimum pressure standard to ensure that users can use water normally, set a corresponding safety margin factor for each low-peak sub-period. The safety margin factor is to add a certain pressure margin on the basis of the benchmark value to prevent the actual pressure from being too low and lower than the minimum service head. Since the difference distribution of different sub-periods is different, that is, the gap between the minimum pressure value and the benchmark value is different, it is necessary to set a different safety margin factor for each sub-period to ensure that the pressure in each sub-period can meet the minimum service head requirement.

[0038] For each off-peak sub-period, its pressure baseline value is multiplied by the safety margin coefficient to obtain the off-peak pressure safety threshold of the sub-period. This is because the safety margin coefficient is set based on the ratio of the baseline value. The threshold obtained after multiplication takes into account the typical pressure level (baseline value) of the sub-period and includes the necessary safety margin, which can ensure that the water supply pressure of the sub-period will not be lower than the minimum service head; the thresholds of all off-peak sub-periods are summarized to form a complete off-peak pressure safety threshold system. Such a system can cover all sub-periods of the entire off-peak period, so that the control system can have corresponding pressure control standards in different off-peak sub-periods, thereby achieving more accurate and reasonable off-peak pressure control.

[0039] S5. Real-time monitoring of the water supply pressure at the terminal pressure control point. When the pressure is lower than the low-peak pressure safety threshold, a pre-alarm is triggered and pressure changes are continuously recorded. If the pre-alarm state persists, the control system switches to the peak-peak boost control strategy. After the water supply pressure at the terminal pressure control point returns to normal, the low-peak pressure safety threshold for that period is automatically corrected.

[0040] In step S5 of this embodiment, the process of automatically correcting the low-peak pressure safety threshold of the period is as follows: When the water supply pressure at the terminal pressure control point rises back above the original low-peak pressure safety threshold, it means that after the previous strategy adjustment, the pressure at this point has reached the originally set safety standard. The continuous stability time reaches the preset pressure stability time to confirm that this rebound is not a short-term fluctuation, but is truly stable above the safety level. If it only briefly exceeds the threshold and then drops, it may indicate that there are still unstable factors in the system and it cannot be regarded as a true return to normal. Only when these two conditions are met at the same time can it be determined that the pressure has returned to normal.

[0041] The complete data sequence of this pre-alarm event is extracted, including: the duration of the pressure below the threshold, which reflects how long the insufficient pressure state lasts. The longer the duration, the greater the scope and extent of the problem; the pressure fluctuation amplitude, which reflects the drastic change in the pressure value during the insufficient pressure period. The greater the fluctuation, the more unstable the system; the rate of pressure recovery after switching to the peak period strategy, which reflects the effectiveness of the peak period strategy in increasing pressure. A fast recovery indicates that the strategy adjustment is timely and effective; the average pressure value and pressure stability time in the first complete low-peak period after returning to normal. These two data can reflect the stable state of the system after returning to normal. The average value reflects the normal pressure level, and the stability time indicates the stability after recovery. These data together constitute a comprehensive record of this abnormal event and provide a detailed basis for subsequent threshold correction.

[0042] The pressure average of the first complete low-peak period after returning to normal is used as the benchmark. This average is the representative pressure value of the first complete period after the system returns to stability, which can reflect the actual pressure level of the current system under normal conditions. The difference between it and the original low-peak pressure safety threshold is calculated based on this. The actual pressure deviation value obtained can accurately reflect the gap between the original threshold and the current actual normal pressure; combined with the maximum amplitude of the pressure below the threshold in this abnormal event, this amplitude reflects the most serious degree of insufficient pressure, and the deviation coefficient is generated through the proportional algorithm. The proportional algorithm can combine the two different dimensions of parameters, the deviation value and the maximum amplitude, so that the deviation coefficient can not only reflect the gap under normal conditions, but also take into account the degree of insufficient pressure under extreme conditions, thereby more comprehensively reflecting the direction and degree of adjustment of the original threshold.

[0043] The dynamic correction factor is determined based on the deviation coefficient. The dynamic correction factor is set according to the size of the deviation coefficient, which determines the adjustment range of the safety margin coefficient. The larger the deviation coefficient, the greater the gap between the original threshold and the actual situation, the larger the correction factor, and the correspondingly larger the adjustment range; the safety margin coefficient of the original low-peak sub-period is adjusted: if the actual pressure deviation value is positive, it means that the actual pressure after recovery is higher than the original threshold, and the original safety margin may be too large, resulting in a high threshold setting, causing energy waste and other problems, so the safety margin coefficient is lowered; if it is a negative value, it means that the actual pressure after recovery is lower than the original threshold, and the original safety margin is insufficient, and water supply safety may not be guaranteed, so the safety margin coefficient is increased. The adjustment range is determined by the dynamic correction factor, so that the adjusted safety margin coefficient can be more in line with the actual situation of the current system.

[0044] The adjusted safety margin coefficient and the original pressure reference value are used to recalculate and obtain the revised pressure safety threshold for the off-peak sub-period. Because the original pressure reference value is a typical pressure level calculated based on historical stable data, it has a certain stability and representativeness. Combined with the adjusted safety margin coefficient, a new threshold that conforms to the current actual situation and retains historical experience can be obtained; the revised threshold meets the constraint condition of not less than the minimum service head of the water supply system. This is to ensure that no matter how the adjustment is made, the most basic water supply pressure requirements can be guaranteed, and the water supply pressure will not be lower than the minimum standard required for normal water use by users due to the correction, thereby ensuring water supply safety.

[0045] Example 2 See Figure 2 A time-divided intelligent pressure regulation management system for secondary water supply, used to implement the time-divided intelligent pressure regulation management method for secondary water supply of the above embodiment, comprising: The water usage period division module divides a day into peak water usage period and low water usage period based on historical water flow data, and determines the start and end time nodes of each period; The terminal pressure control point locking module locks the terminal pressure control point of the water supply system through hydraulic calculation and actual water supply pressure monitoring, and collects the water supply pressure data of the point in real time; Peak boost control module, used to execute peak boost control strategy during peak water consumption period; The low-peak pressure threshold setting module is used to analyze the historical water supply pressure data of the terminal pressure control point during the low-peak water consumption period and set the low-peak pressure safety threshold; based on the threshold, it sends a pressure instruction to the control system to control the output pressure of the water supply equipment according to the threshold; The terminal pressure monitoring and correction module is used to monitor the water supply pressure of the terminal pressure control point in real time. When the pressure is lower than the low-peak pressure safety threshold, it triggers a pre-alarm and continuously records the pressure changes. If the pre-alarm state continues, the control system switches to the peak-peak boost control strategy. After the water supply pressure of the terminal pressure control point returns to normal, the low-peak pressure safety threshold of the period is automatically corrected.

[0046] The above is a detailed description of an embodiment of the present invention. However, the content described is only a preferred embodiment of the present invention and should not be considered to limit the scope of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A time-divided intelligent pressure regulation management method for secondary water supply, characterized in that: The following steps are involved: S1. Based on historical water flow data, divide the day into peak water use period and low water use period, and determine the start and end time nodes of each period; S2. Lock the pressure control point at the end of the water supply system through hydraulic calculation and actual water supply pressure monitoring, and collect water supply pressure data at this point in real time. The process of locking the pressure control point at the end of the water supply system is as follows: S21. Build a digital twin model of the water supply system; S22. Based on the digital twin model, input historical typical operating conditions covering water flow data in different time periods and seasons, conduct steady-state and transient hydraulic simulations using hydraulic calculation software, output the theoretical pressure value of each node, and generate a multi-operating condition theoretical pressure field distribution. Mark the candidate node set with the lowest theoretical pressure under each operating condition. S23. Based on the distribution characteristics of the candidate nodes in the theoretical pressure field, monitoring points are arranged along the pipeline and at the user end; S24. During typical water usage periods, covering both peak and off-peak periods, the digital twin model and the dynamic monitoring network are run simultaneously to collect theoretical and actual pressure data in the same time series. Error analysis is performed by calculating the deviation between the theoretical and actual values. The pipe resistance coefficient and node flow distribution parameters in the model are adjusted until the deviation between the pressure distribution output by the model and the actual monitoring results is within a preset range. S25. Based on the calibrated model and combined with real-time pressure data from the dynamic monitoring network, extract the lowest pressure value of each node throughout the entire operating cycle, the duration of pressure below the preset basic threshold, and the pressure fluctuation amplitude. A comprehensive score is then assigned to all nodes using a multi-feature weighted algorithm. The algorithm weights are set based on the water supply safety priority, and the node with the lowest score is designated as the terminal pressure control point of the water supply system. S3. During the peak water consumption period, the peak pressure boost control strategy is implemented; S4. During the low-peak water consumption period, the historical water supply pressure data of the terminal pressure control point is analyzed to set a low-peak pressure safety threshold; based on the threshold, a pressure instruction is sent to the control system to control the output pressure of the water supply equipment according to the threshold; S5. Real-time monitoring of the water supply pressure at the terminal pressure control point. When the pressure is lower than the low-peak pressure safety threshold, a pre-alarm is triggered and pressure changes are continuously recorded. If the pre-alarm state persists, the control system switches to the peak-peak boost control strategy. After the water supply pressure at the terminal pressure control point returns to normal, the low-peak pressure safety threshold for that period is automatically corrected.

2. The time-divided intelligent pressure regulation management method for secondary water supply according to claim 1 is characterized in that: In step S1, the process of dividing a day into a peak water consumption period and a low water consumption period based on historical water flow data and determining the start and end time nodes of each period is as follows: S11. Collect 30 consecutive days of raw water flow data, convert it into a continuous time series with a time unit of 15 minutes, remove extreme outliers through the box plot method, and use linear interpolation to fill in the missing data segments; S12. Calculate the flow rate value for each time unit, calculate the flow rate average within the window based on a 2-hour sliding window, average the 30-day data at the same time unit, and generate a daily average flow rate distribution curve; S13. Calculate the overall mean and standard deviation based on the daily average traffic distribution curve, delineate a candidate high-peak interval that is higher than the mean by +1.2 times the standard deviation, and a candidate low-peak interval that is lower than the mean by -1.2 times the standard deviation. Iteratively adjust the standard deviation multiple to ensure that the candidate intervals cover more than 90% of the significant fluctuation segments of the curve. S14. Traverse the daily average traffic distribution curve, identify consecutive peak and low-peak candidate time unit clusters and mark them as preliminary time periods, and merge preliminary time periods of the same type with an interval of no more than 1 time unit; S15. Count the frequency of occurrence of each preliminary time period that meets the corresponding traffic characteristics in multi-day data, retain stable time periods with a frequency of not less than 75%, and use the time when three consecutive time units meet the characteristics as the start and end nodes to determine the time range of peak and off-peak periods.

3. The time-divided intelligent pressure regulation management method for secondary water supply according to claim 1 is characterized in that: In step S23, the monitoring points are arranged so that the density of the candidate node set area is higher than that of other areas, and all monitoring points have real-time data transmission capabilities to collect actual operating pressure data of each node.

4. The time-divided intelligent pressure regulation management method for secondary water supply according to claim 1 is characterized in that: In step S3, during the peak water usage period, the peak boost control strategy is executed in the following manner: A fixed peak water supply pressure threshold is preset, and real-time pressure data of the water supply system is obtained through pressure monitoring; the real-time pressure data is compared with the peak water supply pressure threshold. If the real-time pressure is lower than the threshold, the output pressure is increased; if the real-time pressure is higher than the threshold, the output pressure is reduced; according to the comparison results of the real-time pressure data and the peak water supply pressure threshold, corresponding adjustments are made to ensure that the water supply system pressure is always maintained within the peak water supply pressure threshold range.

5. The time-divided intelligent pressure regulation management method for secondary water supply according to claim 1 is characterized in that: In step S4, during the low-peak period of water use, the historical water supply pressure data of the terminal pressure control point is analyzed to set the low-peak pressure safety threshold as follows: S41. Collect the low-peak water supply pressure data of the terminal pressure control points for the past 30 days and classify them by date to form a historical water supply pressure data set of the terminal pressure control points; S42. Based on the determined start and end times of the off-peak period and in combination with historical water flow fluctuation characteristics, the off-peak period is further divided into several off-peak sub-periods; each off-peak sub-period has a fixed duration, and the water flow fluctuation amplitude within each sub-period is within a preset range; S43. For each off-peak period, extract the pressure characteristic parameters of the corresponding period from the historical water supply pressure data set of the terminal pressure control point, including the pressure mean value, pressure minimum value, pressure standard deviation, and continuous stable pressure duration within the sub-period; S44. For each off-peak period, the pressure baseline value is calculated using the weighted average method: S45. Analyze the difference distribution between the historical minimum pressure value and the pressure reference value of each off-peak period, and set a corresponding safety margin factor for each off-peak period in combination with the minimum service head requirement of the water supply system; S46. For each off-peak period, multiply its pressure baseline value by the safety margin coefficient to obtain the off-peak pressure safety threshold of the sub-period; summarize the thresholds of all off-peak sub-periods to form a complete off-peak pressure safety threshold system.

6. The time-divided intelligent pressure regulation management method for secondary water supply according to claim 5, characterized in that: In step S44, it also includes: The pressure average value of the sub-period is weighted by the continuous stable pressure duration in the sub-period; the weight distribution rule is: the pressure average value with a continuous stable duration longer than the preset duration is given a higher weight.

7. The time-divided intelligent pressure regulation management method for secondary water supply according to claim 1 is characterized in that: In step S5, the process of automatically correcting the low-peak pressure safety threshold value of the period is as follows: S51: When the water supply pressure at the terminal pressure control point rises back above the original low-peak pressure safety threshold and the continuous stability time reaches the preset pressure stability time, it is determined that the pressure has returned to normal; S52. Extract the complete data sequence of this pre-alarm event, including: the duration of the pressure below the threshold, the pressure fluctuation amplitude, the rate of pressure recovery after switching to the peak period strategy, the pressure average value in the first complete off-peak period after returning to normal, and the pressure stable duration; S53. Calculate the difference between the mean pressure value of the first complete off-peak period after normal operation and the original off-peak period pressure safety threshold to obtain the actual pressure deviation value. Combined with the maximum magnitude of the pressure below the threshold during this abnormal event, generate a deviation coefficient using a proportional algorithm. S54. Determine a dynamic correction factor based on the deviation coefficient and adjust the safety margin coefficient of the original off-peak period: if the actual pressure deviation value is positive, adjust the safety margin coefficient downward; if it is negative, adjust the safety margin coefficient upward. The adjustment range is determined by the dynamic correction factor. S55. Recalculate the adjusted safety margin coefficient and the original pressure reference value to obtain a revised pressure safety threshold for the off-peak period. The revised threshold satisfies the constraint condition of being no less than the minimum service head of the water supply system.

8. A time-divided intelligent pressure regulation management system for secondary water supply, characterized in that: A time-divided intelligent pressure regulation management method for secondary water supply for implementing any one of claims 1 to 7, comprising: The water usage period division module divides a day into peak water usage period and low water usage period based on historical water flow data, and determines the start and end time nodes of each period; The terminal pressure control point locking module locks the terminal pressure control point of the water supply system through hydraulic calculation and actual water supply pressure monitoring, and collects the water supply pressure data of the point in real time; Peak boost control module, used to execute peak boost control strategy during peak water consumption period; The low-peak pressure threshold setting module is used to analyze the historical water supply pressure data of the terminal pressure control point during the low-peak water consumption period and set the low-peak pressure safety threshold; based on the threshold, it sends a pressure instruction to the control system to control the output pressure of the water supply equipment according to the threshold; The terminal pressure monitoring and correction module is used to monitor the water supply pressure of the terminal pressure control point in real time. When the pressure is lower than the low-peak pressure safety threshold, it triggers a pre-alarm and continuously records the pressure changes. If the pre-alarm state continues, the control system switches to the peak-peak boost control strategy. After the water supply pressure of the terminal pressure control point returns to normal, the low-peak pressure safety threshold of the period is automatically corrected.

Images