Intelligent constant-pressure and pressure-stabilizing method and system for whole-section water supply of living house

Abstract

The invention discloses an intelligent constant-pressure and stable-pressure method and system for whole-section water supply of a living house, and the method comprises the steps: firstly, carrying out constant-pressure and stable-pressure water supply on a municipal direct supply area through the pressure of a municipal water supply pipe; secondly, a secondary water supply pressure stabilizing device is used for meeting the constant-pressure and stable-pressure control requirement for the water supply pressure of the low area and the middle area; thirdly, the pressure value before a pressure reducing valve of a high-area end user is monitored in real time, the outlet pressure of pressurizing equipment is regulated and controlled in real time, and a passive regulation database is established; and finally, after the intelligent constant-pressure and pressure-stabilizing system carries out reinforcement learning on the basis of data of the passive adjustment database, a high-area active adjustment mode is formed, and calibration is carried out through passive adjustment. The strategy can be adaptively optimized, constant pressure and stable pressure of whole-section water supply are guaranteed, meanwhile, system energy consumption is reduced, and the technical problem of pressure hysteresis of a user side is effectively solved.

Description

A smart constant pressure stabilization method and system for water supply throughout residential buildings Technical Field

[0001] This invention relates to the field of constant pressure stabilization technology for secondary water supply, and in particular to an intelligent constant pressure stabilization method and system for the entire water supply system in residential buildings. Background Technology

[0002] Water pressure issues are becoming increasingly prominent in modern residential water supply systems, leading to large pressure fluctuations at high-rise user terminals, water meter idling, reverse rotation, and a high risk of pipe bursts. These problems not only affect residents' daily water usage experience but also increase wear and tear on pipes and equipment, as well as energy consumption.

[0003] In existing technologies, many residential buildings use constant-pressure water supply systems that, while providing a certain degree of pressure stability, still have limitations in practical applications, especially in regulating water pressure at the end of high-rise buildings. Against this backdrop, secondary water supply pressure stabilization systems and intelligent regulation devices have emerged. However, most existing technologies focus on pressure stabilization at the end of the water supply line, neglecting the constant-pressure stabilization process across the entire pipeline. Furthermore, most pressure stabilization technologies are "passive regulation," resulting in pressure regulation lag. For example, patent application CN120465559A, entitled "A Method, Device, and Electronic Equipment for Stabilizing and Saving Energy Control of Pressure Stabilization at the End of a Building Pressurized Water Supply Line," stabilizes the water pressure at the end of the line at a preset threshold, failing to consider constant-pressure stabilization across the entire water supply line and ignoring the initial water pressure fluctuations caused by pressure regulation lag.

[0004] Therefore, there is an urgent need to propose an intelligent constant pressure stabilization system that can effectively achieve constant pressure stabilization and active adjustment of water supply throughout residential buildings. Summary of the Invention

[0005] Based on the above background, this invention provides an intelligent constant pressure stabilization method and system for the entire water supply system in residential buildings. First, it uses the municipal water supply pipe pressure to provide constant pressure stabilization to the municipal direct supply area. Second, it utilizes a secondary water supply pressure stabilization device to meet the constant pressure stabilization control requirements for the low and medium zones. Then, it monitors the pressure values ​​before the pressure reducing valves of end users in the high zones in real time, adjusts the outlet pressure of the pressurization equipment in real time, and establishes a passive adjustment database. During this process, the real-time monitored water pressure values ​​are processed as historical data to build a model to predict future water supply pressure. It is assumed that at time... The monitored water supply pressure is The observed value is then used to adjust the outlet pressure of the pressurizing equipment. Finally, the intelligent constant pressure stabilization system performs reinforcement learning based on the data in the passive adjustment database to form an active adjustment mode for constant pressure stabilization in the high zone. It is then calibrated through passive adjustment to meet the constant pressure stabilization state of the entire water supply in residential buildings, thus solving the technical problems of water pressure fluctuations at the user end and the lag in constant pressure stabilization.

[0006] This invention provides an intelligent constant pressure stabilization method for the entire water supply system in residential buildings. The specific technical solution is as follows:

[0007] A smart constant pressure stabilization method for the entire water supply system in residential buildings includes the following steps:

[0008] S1. Based on the actual water supply situation, the water supply area of ​​each floor is divided into municipal direct supply area, low-pressure area, medium-pressure area and high-pressure area;

[0009] S2. Install secondary water supply pressure stabilizing devices at the most unfavorable points in the low and medium pressurization zones, and install pressure monitoring equipment and flow acquisition equipment at key floor points at the end of the residential buildings to monitor the water supply pressure and flow data at the corresponding points in real time.

[0010] S3. Adjust the outlet pressure of the pressurizing equipment based on the real-time monitoring pressure value at the residential terminal, and collect and store the adjustment data to establish a passive adjustment database.

[0011] S4. Construct an intelligent constant pressure stabilization model based on machine learning algorithms, and train the intelligent constant pressure stabilization model using a passive adjustment database to achieve active adjustment of constant pressure stabilization of water supply in high-rise areas.

[0012] In a further technical solution, in step S2, the pressure monitoring device is arranged in front of the pressure reducing valve at the end of the residence.

[0013] In a further technical solution, the method for adjusting the outlet pressure of the pressurizing equipment in step S3 is as follows:

[0014] Through network communication technology, the water pressure at the end of the secondary water supply system is collected in real time and fed back to the control system of the secondary water supply pressure stabilizing device. The control system adjusts the operating frequency of the water pump in real time according to the pressure of the end user.

[0015] In a further technical solution, in step S3, the adjustment data collected when establishing the passive adjustment database includes the real-time monitoring pressure, equipment outlet pressure, real-time monitoring flow, time, and weather corresponding to the adjustment time.

[0016] In a further technical solution, in step S4, the time series prediction model ARIMA is used as the machine learning algorithm for constructing the intelligent constant voltage regulation model. The expression of the ARIMA model is as follows:

[0017] ARIMA(p, d, q) = AR(p) + I(d) + MA(q), where I(d) is the difference order, AR(p) is the autoregression order, and MA(q) is the moving average order.

[0018] The formulas for the ARIMA model can be specifically expressed as shown in equations (1)-(3) below:

[0019]

[0020] in, This is the current water supply pressure (observed value). , , , Indicates the past Observations at specific points in time (historical data, such as previous water supply pressure); It is the autoregressive coefficient, representing the degree of influence of past data on the current value; This is the error term, representing the portion that the model fails to explain;

[0021]

[0022] in, This is the difference in a time series, representing the change between the current time and the previous time. This part is used to handle trends in the time series and ensure the data is stable.

[0023]

[0024] in, , , , This represents the past error term, indicating the difference between the model prediction and the actual value; , , , This represents the moving average coefficient, indicating the impact of past errors on the current value.

[0025] In a further technical solution, the method for constructing an intelligent constant voltage stabilization model is as follows:

[0026] S41. Data Processing: Handle missing values ​​and outliers in the passive adjustment database, and perform stationarization processing using the ADF test. If the p-value is greater than 0.05, the sequence is considered non-stationary and further differencing is required until the p-value is less than 0.05.

[0027] S42. Model parameters: In Python, the auto_arima function in the pmdarima library is used to automatically traverse and fit all possible combinations of (p, d, q), and determine the optimal parameter combination (p, d, q) of the ARIMA model according to the AIC evaluation criteria, so as to reduce the need for subsequent manual intervention.

[0028] Specifically, the formula for calculating AIC (Akaike Information Criterion) is as follows:

[0029]

[0030] Where k is the number of parameters in the model, and L is the maximum likelihood estimate of the model.

[0031] S43. Model Training: Divide the historical data in the passive conditioning database into 80% training set and 20% validation set, and use the determined optimal parameter combination (p, d, q) to call the ARIMA function to train and fit the model.

[0032] S44. Model Evaluation: After the model training is completed, the model's fit is first evaluated using residual analysis. When the model conforms to a normal distribution and the p-value in the Ljung-Box test is >0.05, the model fit is considered good. Otherwise, it indicates that the model still has room for optimization, and it is necessary to return to the model parameter step, narrow the parameter range, and readjust the combination of (p, d, q). When the model passes the residual analysis, predictions are made on the test set to objectively evaluate the model. The root mean square error (RMSE) is used as the evaluation index. When the RMSE value is <0.85, the model is considered to have good predictive ability for the water supply pressure at the user end.

[0033] Specifically, the formula for calculating RMSE is as follows:

[0034]

[0035] in, This is the actual pressure value. is the predicted stress value, and n is the number of samples in the test set.

[0036] Furthermore, the present invention also provides an intelligent constant voltage stabilizing system based on the above method, the specific technical solution of which is as follows:

[0037] A smart constant pressure stabilizing system for the entire water supply system in residential buildings includes:

[0038] The water supply zone division module is used to divide residential building floors into municipal direct supply zones, low-pressure booster zones, medium-pressure zones, and high-pressure zones, among which:

[0039] Municipal direct supply area: For floors where the municipal water supply pressure can meet the requirements of constant and stable water pressure, the terminal pressure value should be ≥0.14MPa;

[0040] Low zone: The normal pressure is 0.6MPa, suitable for floors of 4 to 14.

[0041] Central zone: The normal pressure is 0.9MPa, suitable for floors of 15 to 25.

[0042] High-rise area: The normal pressure is 1.2MPa, suitable for floors of 26 to 33.

[0043] The secondary water supply pressure stabilizing device module is located at the most unfavorable points in the low and medium pressurization zones to provide secondary water supply pressure stabilization and ensure stable water supply pressure.

[0044] The pressure and flow acquisition module is installed in front of the pressure reducing valve at key floor points at the end of the residential building. It is used to monitor the pressure and flow data of the area in real time and transmit the data to the control system.

[0045] The database module is used to record and store the system's historical regulation data, including real-time monitoring pressure, equipment outlet pressure, real-time monitoring flow, time, and weather data corresponding to the regulation time, and to establish a passive regulation database;

[0046] The ARIMA model prediction module is used to integrate historical adjustment data and predict user-side pressure using the ARIMA model.

[0047] The intelligent constant pressure stabilization module is used to generate control commands and pre-adjust pressurization equipment based on the prediction results of the ARIMA model, so as to realize the active regulation of constant pressure stabilization of water supply in high-rise areas.

[0048] In a further technical solution, the ARIMA model prediction module includes the following sub-modules:

[0049] The data acquisition and preprocessing submodule is used to acquire adjustment data at historical adjustment times and perform noise reduction and differential preprocessing to make it meet the requirements of time series analysis.

[0050] The model training submodule is used to train the model using the ARIMA algorithm, select appropriate p, d, and q parameters, and learn from historical data to capture the changing patterns of water supply pressure.

[0051] The prediction calculation submodule, based on the prediction results of the ARIMA model, predicts the user-end pressure for the next 12 time steps and uses it to adjust the working status of the pressurization equipment, where each time step is 15 minutes.

[0052] In a further technical solution, the intelligent constant voltage regulator module includes the following sub-modules:

[0053] Prediction result generation submodule: Based on the predicted pressure value output by the ARIMA model, it adjusts the operating frequency of the water pump and the working status of the pressurizing equipment in real time through network communication technology to ensure that the water supply pressure can be pre-adjusted to the pressure demand value at the user end.

[0054] The feedback adjustment submodule is used to activate a passive adjustment mechanism when the predicted value deviates from the real-time pressure value of the end user at the subsequent prediction time by more than ±2%, so as to adjust the outlet pressure of the pressurizing equipment to the real-time pressure value.

[0055] Further technical solutions also include:

[0056] The system's self-learning and optimization module is used to continuously adjust the parameters of the ARIMA model based on the adjustment data fed back by the system in real time, thereby improving the accuracy of prediction and the precision of control. During the operation of the system, it further optimizes the ARIMA model by continuously acquiring new pressure and flow data to adapt to changes in different time periods, floors, or external environments.

[0057] Further technical solutions also include:

[0058] The user interface and alarm module provide a user interface to display real-time water supply pressure and flow information, as well as view the system's operating status and pressure distribution. When excessive water pressure fluctuations, equipment malfunctions, or system anomalies occur, alarm information is issued to remind maintenance personnel to inspect and repair the system.

[0059] The beneficial effects of this invention are:

[0060] 1. This invention solves the technical problem of pressure regulation lag in traditional water supply systems by introducing technologies such as secondary water supply pressure stabilization, terminal constant pressure, and machine learning optimization control.

[0061] 2. This invention, through intelligent pressure monitoring and real-time feedback mechanism, coupled with variable frequency drive water pump and secondary water supply pressure stabilization technology, can achieve constant pressure stabilization of the entire water supply system, ensuring stable water pressure in each area (whether low-rise, mid-rise, or high-rise), adapting to different types of buildings, and has the advantage of high-precision water pressure regulation, solving the technical problem of excessive or insufficient water pressure fluctuation in traditional systems.

[0062] 3. The intelligent constant pressure stabilization system built based on the time series model ARIMA in this invention enables the system to respond to water demand in a timely manner, pre-adjust and control the outlet pressure of the pressurization equipment, reduce the lag in water pressure adjustment, and realize "active adjustment" in response to water demand, thereby achieving more intelligent and efficient water supply management and avoiding the pressure regulation lag in traditional water supply systems. Attached Figure Description

[0063] Figure 1 is a flowchart of the method described in this embodiment;

[0064] Figure 2 is a flowchart of the system adjustment process described in this embodiment;

[0065] Figure 3 is a comparison diagram of the user terminal pressure before and after applying the intelligent constant pressure stabilizing system described in the embodiment of the present invention;

[0066] Figure 4 is a diagram showing the operating conditions during peak hours (20:30-22:30) after applying the intelligent constant voltage stabilization system described in this embodiment of the invention;

[0067] Figure 5 is a diagram of the operating conditions during peak hours (20:30-22:30) without applying the intelligent constant voltage stabilization system described in the embodiments of the present invention. Detailed Implementation

[0068] The embodiments of the present invention will be further described below with reference to the accompanying drawings. Embodiment 1:

[0069] A smart constant pressure stabilization method for the entire water supply system in residential buildings, as shown in Figure 1, includes the following steps:

[0070] S1. Based on the actual water supply situation, the water supply area of ​​each floor is divided into municipal direct supply area, low-pressure area, medium-pressure area and high-pressure area;

[0071] S2. Install secondary water supply pressure stabilizing devices at the most unfavorable points in the low and medium pressure zones, and install pressure monitoring equipment and flow acquisition equipment in front of the pressure reducing valves at key floor points at the end of the residential buildings to monitor the water supply pressure and flow data at the corresponding points in real time.

[0072] S3. Through network communication technology, the water pressure at the end of the secondary water supply system is collected in real time and fed back to the control system of the secondary water supply pressure stabilizing device. The control system adjusts the pump operating frequency in real time according to the pressure of the end user, and collects and stores the adjustment data, including the real-time monitoring pressure, equipment outlet pressure, real-time monitoring flow, time, and weather corresponding to the adjustment time, and establishes a passive adjustment database.

[0073] S4. The ARIMA time series forecasting model is used as the machine learning algorithm to construct the intelligent constant pressure stabilization model. The intelligent constant pressure stabilization model is trained using a passive regulation database to achieve active regulation of constant pressure stabilization in high-pressure water supply areas. The expression of the ARIMA model is as follows:

[0074] ARIMA(p, d, q) = AR(p) + I(d) + MA(q), where d is the difference order, p is the autoregressive order, and q is the moving average order.

[0075] The specific steps for constructing an intelligent constant voltage stabilization model are as follows:

[0076] S41. Data Processing: Handle missing values ​​and outliers in the passive adjustment database, and perform stationarization processing using the ADF test. If the p-value is greater than 0.05, the sequence is considered non-stationary and further differencing is required until the p-value is less than 0.05.

[0077] S42. Model parameters: In Python, the auto_arima function in the pmdarima library is used to automatically traverse and fit all possible combinations of (p, d, q), and determine the optimal parameter combination (p, d, q) of the ARIMA model according to the AIC evaluation criteria, so as to reduce the need for subsequent manual intervention.

[0078] S43. Model Training: Divide the historical data in the passive conditioning database into 80% training set and 20% validation set, and use the determined optimal parameter combination (p, d, q) to call the ARIMA function to train and fit the model.

[0079] S44. Model Evaluation: After the model training is completed, the model's fit is first evaluated using residual analysis. When the model conforms to a normal distribution and the p-value in the Ljung-Box test is >0.05, the model fit is considered good. Otherwise, it indicates that the model still has room for optimization, and it is necessary to return to the model parameter step, narrow the parameter range, and readjust the (p, d, q) combination. When the model passes the residual analysis, predictions are made on the test set to objectively evaluate the model. The root mean square error (RMSE) is used as the evaluation index. When the RMSE value is <0.85, the model is considered to have good predictive ability for the water supply pressure at the user end.

[0080] In this embodiment, the system installation process first needs to be planned according to the structural characteristics of the building. During the installation phase, pressure monitors are first deployed on different floors and at key locations in the high-rise building. These monitors can capture real-time water pressure changes in each area. Then, based on actual water pressure requirements, corresponding secondary water supply pressure stabilizing equipment is installed in the low and middle zones to maintain stable water pressure at the most unfavorable points. Pressure and flow monitors continuously monitor data at the high-end terminals to ensure that the water supply system can control the pressure in the high-end areas at any time.

[0081] After installation, the intelligent constant pressure stabilization system is connected to real-time data monitors, pressurization equipment, and other devices via network communication technology to ensure real-time data transmission and adjustment. The intelligent constant pressure stabilization system, trained on a large amount of historical data, predicts water pressure demand for the next 12 time steps, each time step being 15 minutes. The intelligent constant pressure stabilization system transmits the predicted water pressure data for the first time step to the pressurization equipment. The pressurization equipment flexibly adjusts the pump speed based on this data, and the secondary water supply pressure stabilization equipment buffers water pressure fluctuations, thereby achieving "active adjustment" of constant water pressure. The specific adjustment process is shown in Figure 2.

[0082] During actual system operation, the intelligent constant pressure stabilization system provides real-time feedback and adaptive optimization, gradually adjusting the water supply pressure to avoid unnecessary fluctuations. For example, during peak water usage periods, the water pressure may tend to decrease. In this case, the intelligent constant pressure stabilization system unit will increase the pump speed in advance based on the prediction of the time series model to ensure stable water pressure. During off-peak water usage periods, the system will save energy by reducing the pump speed, avoiding energy waste. It is worth noting that when the time reaches the first time step, if the predicted value deviates from the real-time pressure value by more than ±2%, "passive adjustment" is used for calibration. That is, the intelligent constant pressure stabilization system transmits the real-time pressure value to the pressurization equipment, which then makes real-time adjustments. Energy consumption data from a pilot application in a certain community shows that after using the technology of this invention, energy consumption decreased by approximately 12.46%.

[0083] As shown in Figures 3-5, the entire system's workflow relies on real-time data feedback and intelligent adjustment to ensure that the water supply pressure in each area remains within the optimal range, preventing excessively high or low pressure. With the intelligent constant pressure stabilization system, the pressurization equipment automatically adjusts in advance based on predicted user water demand and is calibrated according to real-time demand, achieving "variable pressure and variable flow water supply." This ensures sufficient pressure at the user end while automatically optimizing equipment pressure. Furthermore, with the assistance of secondary water supply pressure stabilization equipment, it effectively reduces rapid changes and large fluctuations in terminal pressure. During peak hours (20:30-22:30), the pressurization equipment operates very regularly, with stable water pressure at the end users and the equipment regularly entering a dormant state. However, when not in use, the operation is irregular, with unstable water pressure at the end users and the equipment almost never entering a dormant state.

[0084] Finally, the intelligent constant pressure stabilization system also features anomaly detection and fault prediction functions. By analyzing historical data accumulated during operation, the system can identify potential fault signs. For example, if the water pressure in a section of the pipeline changes beyond the normal range, or if the water pump operates abnormally, the system will issue an early warning, reminding staff to check the equipment. Through this predictive maintenance, the system can intervene before problems occur, avoiding downtime and repair costs caused by equipment failure. Example 2:

[0085] A smart constant pressure stabilizing system for the entire water supply system in residential buildings includes:

[0086] The water supply zone division module is used to divide residential building floors into municipal direct supply zones, low-pressure booster zones, medium-pressure zones, and high-pressure zones, among which:

[0087] Municipal direct supply area: For floors where the municipal water supply pressure can meet the requirements of constant and stable water pressure, the terminal pressure value should be ≥0.14MPa;

[0088] Low zone: The normal pressure is 0.6MPa, suitable for floors of 4 to 14.

[0089] Central zone: The normal pressure is 0.9MPa, suitable for floors of 15 to 25.

[0090] High-rise area: The normal pressure is 1.2MPa, suitable for floors of 26 to 33.

[0091] The secondary water supply pressure stabilizing device module is located at the most unfavorable points in the low and medium pressurization zones to provide secondary water supply pressure stabilization and ensure stable water supply pressure.

[0092] The pressure and flow acquisition module is installed in front of the pressure reducing valve at key floor points at the end of the residential building. It is used to monitor the pressure and flow data of the area in real time and transmit the data to the control system.

[0093] The database module is used to record and store the system's historical regulation data, including real-time monitoring pressure, equipment outlet pressure, real-time monitoring flow, time, and weather data corresponding to the regulation time, and to establish a passive regulation database;

[0094] The ARIMA model prediction module is used to integrate historical adjustment data and predict user-side pressure using the ARIMA model.

[0095] The intelligent constant pressure stabilization module is used to generate control commands and pre-adjust pressurization equipment based on the prediction results of the ARIMA model, so as to realize the active adjustment of constant pressure stabilization of water supply in high-temperature areas.

[0096] The system's self-learning and optimization module is used to continuously adjust the parameters of the ARIMA model based on the adjustment data fed back by the system in real time, thereby improving the accuracy of prediction and the precision of control. During the operation of the system, it further optimizes the ARIMA model by continuously acquiring new pressure and flow data to adapt to changes in different time periods, floors, or external environments.

[0097] The user interface and alarm module provide a user interface to display real-time water supply pressure and flow information, as well as view the system's operating status and pressure distribution. When excessive water pressure fluctuations, equipment malfunctions, or system anomalies occur, alarm information is issued to remind maintenance personnel to inspect and repair the system.

[0098] In this embodiment, the ARIMA model prediction module includes the following sub-modules:

[0099] The data acquisition and preprocessing submodule is used to acquire adjustment data at historical adjustment times and perform noise reduction and differential preprocessing to make it meet the requirements of time series analysis.

[0100] The model training submodule is used to train the model using the ARIMA algorithm, select appropriate p, d, and q parameters, and learn from historical data to capture the changing patterns of water supply pressure.

[0101] The prediction calculation submodule, based on the prediction results of the ARIMA model, predicts the user-end pressure for the next 12 time steps and uses it to adjust the working status of the pressurization equipment, where each time step is 15 minutes.

[0102] In this embodiment, the intelligent constant voltage regulator module includes the following sub-modules:

[0103] Prediction result generation submodule: Based on the predicted pressure value output by the ARIMA model, it adjusts the operating frequency of the water pump and the working status of the pressurizing equipment in real time through network communication technology to ensure that the water supply pressure can be pre-adjusted to the pressure demand value at the user end.

[0104] The feedback adjustment submodule is used to activate a passive adjustment mechanism when the predicted value deviates from the real-time pressure value of the end user at the subsequent prediction time by more than ±2%, so as to adjust the outlet pressure of the pressurizing equipment to the real-time pressure value.

[0105] The embodiments described above are merely illustrative of specific implementations of the present invention, and while the descriptions are detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention.

Claims

1. A smart constant pressure stabilization method for the entire water supply system in residential buildings, characterized in that, Includes the following steps: S1. Based on the actual water supply situation, the water supply area of ​​each floor is divided into municipal direct supply area, low-pressure boosting area, medium-pressure area, and high-pressure area; S2. Secondary water supply pressure stabilizing devices are installed at the most unfavorable points in the low-pressure boosting area and medium-pressure area, and pressure monitoring equipment and flow acquisition equipment are installed at key floor points at the end of the residential building to monitor the water supply pressure and flow data at the corresponding points in real time; S3. The outlet pressure of the boosting equipment is adjusted according to the real-time monitoring pressure value at the end of the residential building, and the adjustment data is collected and stored to establish a passive adjustment database; S4. Construct an intelligent constant pressure stabilization model based on machine learning algorithms, and train the intelligent constant pressure stabilization model using a passive adjustment database to achieve active adjustment of constant pressure stabilization of water supply in high-rise areas.

2. The intelligent constant pressure stabilization method for the entire water supply system in residential buildings according to claim 1, characterized in that, In step S3, the method for adjusting the outlet pressure of the pressurizing equipment is as follows: the water pressure at the end of the secondary water supply system is collected in real time through network communication technology and fed back to the control system of the secondary water supply pressure stabilizing device. The control system adjusts the operating frequency of the water pump in real time according to the pressure of the end user.

3. The intelligent constant pressure stabilization method for the entire water supply system in residential buildings according to claim 1, characterized in that, In step S3, the adjustment data collected when establishing the passive adjustment database includes the real-time monitoring pressure, equipment outlet pressure, real-time monitoring flow rate, time, and weather corresponding to the adjustment time.

4. The intelligent constant pressure stabilization method for the entire water supply system in residential buildings according to claim 1, characterized in that, In step S4, the time series prediction model ARIMA is used as the machine learning algorithm to construct the intelligent constant voltage stabilization model. The expression of the ARIMA model is as follows: ARIMA(p, d, q) = AR(p) + I(d) + MA(q), where d is the difference order, p is the autoregression order, and q is the moving average order.

5. The intelligent constant pressure stabilization method for the entire water supply system in residential buildings according to claim 4, characterized in that, The method for constructing an intelligent constant voltage regulation model is as follows: S41, Data Processing: Missing values ​​and outliers are processed in the passive regulation database. The data is then stationary using the ADF test. If the p-value is greater than 0.05, the sequence is considered non-stationary and further differencing is required until the p-value is less than 0.

05. S42, Model Parameters: In Python, the `auto_arima` function from the `pmdarima` library is used to automatically iterate and fit all possible combinations of (p, d, q). The optimal parameter combination (p, d, q) for the ARIMA model is determined according to the AIC evaluation criteria to reduce subsequent manual intervention. S43, Model Training: The historical data in the passive regulation database is divided into 80%... The training set and a 20% validation set are used to train and fit the model using the determined optimal parameter combination (p, d, q) by calling the ARIMA function. S44. Model Evaluation: After model training, the model's fit is first evaluated using residual analysis. If the model conforms to a normal distribution and the p-value in the Ljung-Box test is >0.05, the model fit is considered good. Otherwise, the model has room for optimization, requiring a return to the model parameter step to narrow the parameter range and readjust the (p, d, q) combination. When the model passes the residual analysis, predictions are made on the test set to objectively evaluate the model. The root mean square error (RMSE) is used as the evaluation metric. When the RMSE value is <0.85, the model is considered to have good predictive ability for the water supply pressure at the user end.

6. An intelligent constant pressure stabilizing system for the entire water supply system in residential buildings, characterized in that, include: The system comprises four modules: a water supply zone division module, a secondary water supply pressure stabilization module, and a secondary water supply pressure monitoring module. The secondary water supply pressure stabilization module is located at the most unfavorable points in the low and medium pressure zones to provide stable secondary water supply pressure. The pressure and flow acquisition module is located before pressure reducing valves at key floor levels in the residential buildings to monitor pressure and flow data in real time and transmit the data to the control system. The database module records and stores historical system regulation data, including real-time monitored pressure, equipment outlet pressure, real-time monitored flow, time, and weather data at the time of regulation, establishing a passive regulation database. The ARIMA model prediction module integrates historical regulation data and uses the ARIMA model to predict user-end pressure. The intelligent constant pressure stabilization module is used to generate control commands and pre-adjust pressurization equipment based on the prediction results of the ARIMA model, so as to realize the active regulation of constant pressure stabilization of water supply in high-rise areas.

7. The intelligent constant pressure stabilizing system for the entire water supply system in residential buildings according to claim 6, characterized in that, The ARIMA model prediction module includes the following sub-modules: a data acquisition and preprocessing sub-module, used to acquire adjustment data at historical adjustment times and perform noise reduction and differential preprocessing to meet the requirements of time series analysis; a model training sub-module, used to train the model using the ARIMA algorithm, select appropriate p, d, and q parameters, and learn from historical data to capture the changing patterns of water supply pressure; and a prediction calculation sub-module, based on the prediction results of the ARIMA model, to predict the user-end pressure for the next 12 time steps and to adjust the working status of the pressurization equipment, wherein each time step is 15 minutes.

8. The intelligent constant pressure stabilizing system for the entire water supply system in residential buildings according to claim 6, characterized in that, The intelligent constant pressure stabilization module includes the following sub-modules: Prediction result generation sub-module: used to adjust the operating frequency of the water pump and the working status of the pressurizing equipment in real time through network communication technology based on the predicted pressure value output by the ARIMA model, so as to ensure that the water supply pressure can be pre-adjusted to the pressure demand value at the user end; The feedback adjustment submodule is used to activate a passive adjustment mechanism when the predicted value deviates from the real-time pressure value of the end user at the subsequent prediction time by more than ±2%, so as to adjust the outlet pressure of the pressurizing equipment to the real-time pressure value.

9. The intelligent constant pressure stabilizing system for the entire water supply system in residential buildings according to claim 6, characterized in that, Also includes: The system's self-learning and optimization module is used to continuously adjust the parameters of the ARIMA model based on the adjustment data fed back by the system in real time, thereby improving the accuracy of prediction and the precision of control. During the operation of the system, it further optimizes the ARIMA model by continuously acquiring new pressure and flow data to adapt to changes in different time periods, floors, or external environments.

10. The intelligent constant pressure stabilizing system for the entire water supply system in residential buildings according to claim 6, characterized in that, Also includes: The user interface and alarm module provide a user interface to display real-time water supply pressure and flow information, as well as view the system's operating status and pressure distribution. When excessive water pressure fluctuations, equipment malfunctions, or system anomalies occur, alarm information is issued to remind maintenance personnel to inspect and repair the system.

Images