Secondary water supply end user pressure stabilization energy-saving control device and method based on water consumption real-time prediction

Abstract

The invention discloses a secondary water supply end user pressure-stabilizing and energy-saving control device and method based on water consumption real-time prediction, and belongs to the field of secondary water supply intelligent control, and the device comprises an Internet of Things sensing system, a water consumption information prediction system, a secondary water supply pipe network digital simulation system, a pump set intelligent control system and a computer management and cloud platform system. Pipeline models of different buildings are obtained through rapid modeling, secondary water supply pipeline head loss values under different flow conditions are obtained by combining pressure sensing correction of the most unfavorable point based on water consumption prediction calculation, and then the pump set outlet pressure which should be met in real time under the condition that the user terminal pressure is constant is obtained. And the frequency of the intelligent control system of the pump set is dynamically adjusted, so that the outlet pressure of the water pump is optimally controlled, different water supply demands are met, frequent changes of pressure of end users are reduced to the maximum extent, and the power consumption of the pump set is remarkably reduced while the water use experience of residents is improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control of secondary water supply, and particularly to a voltage stabilization and energy saving control device and method for end-users of secondary water supply based on real-time prediction of water consumption. Background Art

[0002] As an important part of the municipal water supply system, secondary water supply is characterized by relatively independent pipe networks and very high operating energy consumption. The core of which is the energy consumption of the water pump unit, accounting for more than 95% of the secondary water supply energy consumption. Since there has been a lack of data on the variation law of water consumption in various buildings in China, when designing a building variable frequency speed regulation water supply system, designers usually select pumps according to the "maximum design flow rate" of the system as required by the "Design Standard for Building Water Supply and Drainage". When calculating the "maximum design flow rate" using empirical formulas, although the safety of water supply is considered, in actual operation, due to the overcapacity of the water pump, unreasonable equipment configuration and operation mode, the calculation result is often much larger than the maximum instantaneous flow rate during the actual operation of the system. The system demand does not match the best working condition of the water pump for a long time, which not only reduces the efficiency of the water pump, but also affects the reliability and service life of the water pump. This is because the mainstream variable frequency speed regulation water supply pressure feedback point is usually set near the pump outlet. This water supply method mainly meets the requirements of the maximum flow rate and head of the household, but ignores the water consumption variation law of the household. Especially during the low peak water consumption at night, the pipe resistance will decrease with the change of the pressure in the pipe, resulting in redundant head and energy consumption.

[0003] To solve this problem, a water supply method with constant pressure at the user end is proposed. While meeting the hydraulic requirements of users, it further develops the energy saving of the water supply system, reduces the energy consumption surplus caused by the change of pipe resistance due to the change of flow rate, and thus realizes an ideal energy consumption state of the water supply system. Although relevant enterprises have accumulated certain data on the actual building water consumption, the in-depth analysis and mining of the data are still insufficient. At the same time, there is a lack of comprehensive summary of universal water consumption laws and real-time water consumption prediction, which makes the relevant data unable to provide more accurate feedback and guidance for water pump design and operation.

[0004] Existing patents mainly focus on deploying sensors and corresponding devices in actual buildings to achieve real-time monitoring of the flow rate and pressure at the least utilized water points, and then achieve constant pressure control at the user end. However, since a large number of sensors need to be deployed in actual operating buildings for such devices, the sensor test data often fluctuates greatly during real-time transmission, and the signal attenuation is obvious, failing to meet the accuracy requirements, which has great limitations in large community applications. For example, analyzing an existing invention patent, a pump group intelligent control system with an end pressure monitoring function, mainly regulates the pump group through the end pressure monitoring data of multiple pressure sensors. It uses relevant indicators such as the water consumption characteristics of the community, water consumption prediction, and the required outflow pressure at the most unfavorable point for control calculation. However, the main purpose of its built-in AI algorithm is to intelligently fill in the required pressure values for water supply during missing periods, rather than predicting the water consumption at the next moment. Therefore, in order to achieve the purpose of constant pressure at the user end, it still relies on multiple pressure sensors installed at the most unfavorable point to regulate the operation of the pump group in real time. Summary of the Invention

[0005] To make up for the deficiencies of the existing technology, the present invention provides a secondary water supply end user voltage stabilization and energy-saving control device and method based on water consumption prediction. On the premise of real-time prediction of water consumption in different buildings, it obtains the pipeline head loss corresponding to different flow rates at different times of the day and the outlet pressure of the pump group that should be satisfied when the user end pressure is constant. Only by installing a pressure sensor at the most unfavorable control point of the highest floor pipe network end, the frequency of the pump group intelligent control system can be dynamically adjusted through the system to meet the water supply requirements of different buildings at different times and minimize the frequent changes in the end user pressure. At the same time, the computer management system and cloud platform can be used to analyze and remotely adjust the data, improving the existing ordinary variable frequency constant pressure water supply technology.

[0006] To achieve the above object, the present invention adopts the following technical solutions:

[0007] In the first aspect, the present invention provides a secondary water supply end user voltage stabilization and energy-saving control device based on water consumption prediction, including:

[0008] An Internet of Things perception system that real-time monitors and collects data on the operating frequency, operating power consumption, pump group outlet flow rate, pump group outlet pressure, water tank liquid level, and the pressure at the most unfavorable control point of the highest floor pipe network end of the water pump;

[0009] A water consumption information prediction system that, based on the data collected by the Internet of Things perception system, collects building information through a historical water consumption information collection module and establishes a secondary water supply pipeline layout template, and predicts the water consumption of the building through a water consumption time series prediction module;

[0010] The secondary water supply pipe network digital simulation system predicts the prediction results of the system according to the water usage information, generates the internal building water supply pipe network topology model through the water supply pipe network topology structure rapid modeling module, and calculates the head loss at different flow rates and the calculated value of the pump group outlet pressure through the pipe network head loss probability calculation module;

[0011] The pump group intelligent control system dynamically adjusts the operating frequency of the pump group through the PLC frequency conversion controller according to the calculated value of the pump group outlet pressure calculated by the secondary water supply pipe network digital simulation system, so that the measured value of the pump group outlet pressure is stabilized within the range of 100% - 120% of the calculated value of the pump group outlet pressure;

[0012] The computer management and cloud platform system includes an edge computing module and a human-computer interaction interface. The edge computing module is used to store, calculate and analyze, handle anomalies and push remote data for the data obtained from the water usage information prediction system and the secondary water supply pipe network digital simulation system. The human-computer interaction interface is used to display system information and control settings.

[0013] In a preferred embodiment, the Internet of Things perception system further includes a wired connection device for transmitting the collected data of the pump group outlet flow rate, pressure, water tank liquid level and power consumption to the PLC controller, and the PLC controller transmits it to the computer management and cloud platform system through remote communication.

[0014] In a preferred embodiment, the historical water usage information collection module of the water usage information prediction system includes:

[0015] The basic information collection sub-module is used to collect basic information including water supply area, floor area, floor height, and pipeline routing;

[0016] The key information collection sub-module is used to collect key information for independent water supply systems including water tanks (cisterns), electric control devices, and pressure water containers;

[0017] The secondary water supply pipeline layout template library construction sub-module is used to establish a secondary water supply pipeline layout template library for different types of building floors according to the collected basic information and key information, and provide tools for editing floor templates, supporting users to dynamically edit and supplement the templates according to the floor height and the distribution of water-using appliances;

[0018] The water consumption unit database construction sub-module is used to design the water consumption unit according to the water-using appliances in the building's water supply and drainage, and construct a water consumption unit database for different water-using appliances in combination with the actual seasonal and north-south regional differences and the historical data of the water usage of actual water-using appliances.

[0019] In a preferred embodiment, the water consumption time series prediction module of the water consumption information prediction system includes: a data filtering sub-module, an enhanced empirical mode decomposition (EEMD) sub-module, and a long short-term memory network (LSTM) sub-module;

[0020] The data filtering sub-module is used to receive the real-time monitoring data in the Internet of Things perception system, perform filtering processing using data filtering technology, and correct all real-time flow monitoring data;

[0021] The EEMD sub-module is used to perform intrinsic mode function decomposition on the flow data signal processed by the data filtering sub-module, and use the enhanced empirical mode decomposition (EEMD) technology to decompose the flow data signal into intrinsic mode functions;

[0022] The LSTM sub-module is used to train the water consumption time series data decomposed by the EEMD sub-module, capture the interdependence between each frequency data and time using the deep learning algorithm of the long short-term memory network (LSTM), predict and export the building water consumption data, and form a big data sample of water usage habits, that is, a 24-hour flow time curve table.

[0023] In a preferred embodiment, the rapid modeling module of the water supply pipeline topological structure of the secondary water supply pipe network digital simulation system includes: a case information import sub-module and a topological structure generation sub-module;

[0024] The case information import sub-module is used to receive and import the secondary water supply pipeline information of the case building collected by the historical water consumption information collection module in the water consumption information prediction system (including key water supply facilities such as water supply area, floor height, pipeline routes of risers and branch pipes, pipe material types, types and distributions of water-using appliances, specific configurations of water supply pump stations, and regulating water tanks);

[0025] The topological structure generation sub-module is used to quickly generate a topological structure model of the riser and the internal building water supply pipeline based on the detailed information imported by the case information import sub-module.

[0026] In a preferred embodiment, the pipeline head loss probability calculation module of the secondary water supply pipe network digital simulation system includes: a random allocation sub-module, a confidence interval calculation sub-module, a target head loss value calculation sub-module, and a pump group outlet pressure calculation sub-module;

[0027] The random allocation sub-module is used to receive and process the classified summary information of the internal building water-using appliances, randomly allocate the water consumption of all internal building water-using appliances using the Poisson random allocation principle, and generate a pipeline flow and pressure data set;

[0028] The confidence interval calculation sub-module calculates the maximum and minimum values of the head loss at different flows based on the data set of the random allocation sub-module, and sets the confidence interval.

[0029] The target head loss value calculation sub-module is used to calculate the target head loss value within the confidence interval.

[0030] The pump group outlet pressure calculation sub-module comprehensively calculates the pump group outlet pressure value by combining the target head loss value output by the target head loss value calculation sub-module, the preset constant pressure setting value at the user end, and the water tank liquid level value transmitted in real time by the Internet of Things perception system. The specific calculation process is as follows: add the target head loss value to the constant pressure setting value at the user end, and then subtract the measured value of the water tank liquid level from this sum to obtain the calculated value of the pump group outlet pressure.

[0031] In a preferred embodiment, the intelligent control system of the pump group includes a PLC frequency conversion controller and a pump group frequency converter. The PLC frequency conversion controller is used to control and adjust the pump group frequency converter according to the measured value of the pump group outlet pressure and the calculated value of the pump group outlet pressure, so that the measured value of the pump group outlet pressure is maintained within the preset range of the calculated value of the pump group outlet pressure.

[0032] In a preferred embodiment, the edge computing module of the computer management and cloud platform system integrates a remote data push function and supports synchronously pushing the data processed by the edge computing module to the cloud, the client, and the mobile terminal through remote communication technology.

[0033] In a preferred embodiment, the human-computer interaction interface of the computer management and cloud platform system supports displaying basic information of the community, building pipeline models, pump group performance curves, real-time monitoring data, pump group alarm settings, system control settings, and system control logs to users.

[0034] In a second aspect, the present invention provides a secondary water supply end user voltage stabilization and energy saving control method based on water consumption prediction. Based on the secondary water supply end user voltage stabilization and energy saving control device described in the first aspect, the control method includes the following steps:

[0035] S1, real-time monitoring and collecting data on the operating frequency of the water pump, operating power consumption, pump group outlet flow, pump group outlet pressure, water tank liquid level, and the pressure at the most unfavorable control point at the end of the pipe network on the highest floor through the Internet of Things perception system.

[0036] S2, using the water consumption information prediction system, based on historical water consumption data and real-time monitoring data, decomposing the water consumption signal into intrinsic mode functions through the enhanced empirical mode decomposition (EEMD) technology, and training time series data using the long short-term memory network (LSTM) to predict the water consumption of the building.

[0037] S3. The digital simulation system of the secondary water supply network establishes a topological structure model of the internal building water supply pipeline according to the predicted water consumption, simulates the random distribution of water consumption based on the Poisson random distribution principle, calculates the confidence interval of the head loss at different flow rates, and determines the target head loss value;

[0038] S4. Add the target head loss value to the preset constant pressure setting value at the user end and subtract the measured value of the water tank liquid level to obtain the calculated value of the pump group outlet pressure;

[0039] S5. The intelligent control system of the pump group dynamically adjusts the operating frequency of the pump group through the PLC frequency conversion controller, so that the measured value of the pump group outlet pressure is stabilized within the range of 100% - 120% of the calculated value of the pump group outlet pressure.

[0040] In a preferred embodiment, step S2 specifically includes the following steps:

[0041] The flow sensor collects flow data in real time at a preset frequency;

[0042] Adopt data filtering technology to filter the collected flow data;

[0043] The data corrected by the filtering technology is input into the EEMD sub-module;

[0044] Input the decomposed intrinsic mode functions into the LSTM network for training, and output the time series of water consumption in the next 24 hours.

[0045] In a preferred embodiment, in step S3, the method for establishing the topological structure model of the internal building water supply pipeline includes the following steps:

[0046] Import the secondary water supply pipeline information of the case building collected by the historical water consumption information collection module in the water consumption information prediction system (including key water supply facilities such as water supply area, floor height, pipeline routing of risers and branch pipes, pipe material type, type and distribution of water using appliances, specific configuration of the water supply pump station, and regulating water tank, etc.);

[0047] Based on the imported secondary water supply pipeline information of the case building, automatically generate a topological structure model of the internal building water supply pipeline including risers, branch pipes and water using appliance nodes.

[0048] In a preferred embodiment, in step S3, the calculation steps of the head loss confidence interval and the target head loss value include:

[0049] The pipeline head loss probability calculation module sets the maximum water consumption of each type of building as the upper limit of the available water consumption for distribution according to relevant design standards, and sets the water consumption of the smallest water using appliance as the lower limit of the available water consumption for distribution;

[0050] Classify and summarize all water-using appliances inside the building, sort them from the smallest total water volume to the largest, and adopt the Poisson random distribution principle. Randomly distribute the total available water volume to the water consumption of the actually existing water-using appliances in the building in a way that the water equivalent increase of the water-using appliances in the building gradually increases, so as to obtain the head loss of the topological structure model of the internal water supply pipeline of the building under different flow rates; wherein, the total number of distribution calculations of the Poisson random distribution is 0.3 - 0.5m 3 / (h·10,000 times), and each time a ΔQ water volume is superimposed during distribution to generate a pipeline flow rate - pressure data set;

[0051] Take 20% of the maximum head loss value of the topological structure model of the internal water supply pipeline of the building under different flow rates as the credible minimum head loss value, and 95% as the credible maximum head loss value;

[0052] Calculate the average value of the minimum head loss value and the maximum head loss value, and set this average value as the target head loss value.

[0053] In a preferred embodiment, step S4 specifically includes the following steps:

[0054] According to the flow rate time curve obtained by the water consumption information prediction system, at the predicted flow rate corresponding to a certain time, the pipeline head loss probability calculation module of the secondary water supply network digital simulation system obtains the target head loss value at this flow rate;

[0055] Add the target head loss value at this flow rate to the preset constant pressure setting value of the user terminal to obtain the outlet pressure of each regional distribution pipe;

[0056] Subtract the water tank liquid level value transmitted by the Internet of Things perception system from the outlet pressure of each regional distribution pipe to obtain the required pump head at this flow rate, and further obtain the calculated value of the pump group outlet pressure.

[0057] In a preferred embodiment, the method further includes: performing parameter checking on the topological structure model of the internal water supply pipeline of the building established by the secondary water supply network digital simulation system, specifically including the following steps:

[0058] a) Compare the pressure value of the most unfavorable control point at the end of the highest floor pipe network measured by the Internet of Things perception system with the end pressure value simulated and calculated by the topological structure model of the internal water supply pipeline of the building;

[0059] b) According to the comparison result, dynamically adjust the pipe friction coefficient and the local resistance coefficient, and recalculate the head loss of the topological structure model of the internal water supply pipeline of the building and the pump group outlet pressure;

[0060] c) Repeat a) and b) until the error between the calculated value of the pump group outlet pressure and the measured value of the pump group outlet pressure is stabilized within 5% of the floor height.

[0061] In a more preferred embodiment, the dynamic adjustment of the pipeline friction coefficient and the local resistance coefficient includes: adjusting the pipeline friction coefficient within the range of Hazen-Williams coefficient from 90 to 130, and simultaneously adjusting the local resistance coefficient within the range of 0.1 to 10.0.

[0062] In a more preferred embodiment, the method further includes: the edge computing module of the computer management and cloud platform system analyzes the energy consumption of the pump group, including the following steps:

[0063] Receiving and recording the time data of the pump group operation and the corresponding pump group power consumption data;

[0064] Processing the instantaneous power consumption data and the cumulative power consumption data of the pump group based on the actual operation conditions of the pump group;

[0065] Generating and outputting the relationship curve between the operation time of the pump group and the cumulative power consumption according to the processing results.

[0066] Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

[0067] The present invention discloses a secondary water supply end-user voltage stabilization and energy-saving control device and method based on real-time water consumption prediction, belonging to the field of intelligent control of secondary water supply, including an Internet of Things perception system, a water consumption information prediction system, a digital simulation system of the secondary water supply network, a pump group intelligent control system, and a computer management and cloud platform system. The present invention obtains the pipeline models of different buildings through rapid modeling, and based on the water consumption prediction calculation and the pressure sensing correction of the most unfavorable point, obtains the head loss values of the secondary water supply pipeline under different flow conditions, and then obtains the outlet pressure of the pump group that should be satisfied in real time under the condition of constant pressure at the user end, dynamically adjusts the frequency of the pump group intelligent control system, thereby optimizing the control of the pump outlet pressure, meeting different water supply demands and minimizing the frequent changes in the pressure of the end users, while improving the residents' water use experience and significantly reducing the power consumption of the pump group. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0069] Figure 1 It is the control logic diagram of the secondary water supply end-user voltage stabilization and energy-saving control method based on water consumption prediction of the present invention;

[0070] Figure 2It is a comparison chart of the water consumption of the project before and after the transformation in the implementation case of the present invention;

[0071] Figure 3 It is a comparison chart of the power consumption of the project before and after the transformation in the implementation case of the present invention;

[0072] Figure 4 It is a comparison chart of the energy consumption per thousand tons of water of the project before and after the transformation in the implementation case of the present invention. Detailed implementation manners

[0073] In order to make the above and other features and advantages of the present invention clearer, the present invention will be further described below with reference to the accompanying drawings. It should be understood that the specific embodiments given herein are for the purpose of explaining to those skilled in the art and are merely exemplary, not restrictive.

[0074] In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one of such features. In the description of the present invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise specifically defined.

[0075] Embodiment 1:

[0076] Refer to Figure 1 As shown, this embodiment provides a secondary water supply end-user voltage stabilization and energy-saving control device based on water consumption prediction, including: an Internet of Things perception system, a water consumption information prediction system, a digital simulation system for the secondary water supply pipe network, a pump group intelligent control system, and a computer management and cloud platform system. Among them, the water consumption information prediction system includes a historical water consumption information collection module and a water consumption time series prediction module; the digital simulation system for the secondary water supply pipe network includes a rapid modeling module for the topological structure of the water supply pipeline and a probability calculation module for the head loss of the pipeline; the computer management and cloud platform system includes an edge computing module and a human-computer interaction interface.

[0077] The Internet of Things perception system monitors and collects in real time the data of the operating frequency, operating power consumption, outlet flow of the pump group, outlet pressure of the pump group, water tank liquid level, and the pressure at the most unfavorable control point at the end of the pipe network on the highest floor.

[0078] The water consumption information prediction system, based on the data collected by the Internet of Things perception system, collects building information through the historical water consumption information collection module and establishes a layout template for the secondary water supply pipeline, and predicts the water consumption of the building through the water consumption time series prediction module.

[0079] The secondary water supply pipe network digital simulation system predicts the prediction results of the system based on water usage information, generates the building internal water supply pipe network topology model through the water supply pipe network topology rapid modeling module, and calculates the head loss and the calculated value of the pump group outlet pressure at different flow rates through the pipe network head loss probability calculation module.

[0080] The pump group intelligent control system dynamically adjusts the operating frequency of the pump group through the PLC frequency conversion controller according to the calculated value of the pump group outlet pressure calculated by the secondary water supply pipe network digital simulation system, so that the measured value of the pump group outlet pressure is stabilized within the range of 100% - 120% of the calculated value of the pump group outlet pressure.

[0081] The computer management and cloud platform system includes an edge computing module and a human-computer interaction interface. The edge computing module is used to store, calculate and analyze, handle anomalies, and push remote data for the data obtained from the water usage information prediction system and the secondary water supply pipe network digital simulation system. The human-computer interaction interface is used to display system information and control settings.

[0082] The above systems will be introduced in detail as follows:

[0083] 1. Internet of Things perception system

[0084] The Internet of Things perception system can monitor the pump operating frequency, operating power consumption, pump group outlet flow rate, pump group outlet pressure, water tank liquid level, and the most unfavorable control point pressure at the end of the pipe network on the highest floor, and collect and store these monitored data in real time. The pump group outlet flow rate, pressure, water tank liquid level, and power consumption data are connected to the PLC controller through wire, and the PLC controller transmits them to the computer management and cloud platform system through remote communication for subsequent analysis and processing.

[0085] 2. Water usage information prediction system

[0086] The water usage information prediction system includes a historical water usage information collection module and a water usage time series prediction module.

[0087] The historical water usage information collection module can collect basic information including water supply area, floor area, floor height, pipe routing, etc. The independent water supply system composed of facilities such as water tanks (cisterns), electric control devices, and pressure water containers should be collected as key information. According to the collected information, a secondary water supply pipe layout template library for different types of building floors is established, and tools for editing floor templates are provided. For similar buildings with different floor heights and different water-using appliances, it can be supplemented and improved on the basis of the secondary water supply pipe layout template library. At the same time, this module can design the water consumption per unit for water-using appliances according to the water-using appliances in the building's water supply and drainage, and construct a water consumption per unit database for different water-using appliances in combination with the actual seasonal and north-south regional differences and the historical data of the actual water usage of actual water-using appliances.

[0088] The water consumption time series prediction module filters the real-time monitoring data in the Internet of Things perception system through local sensors, collects the data with higher frequencies (such as collecting one data every 6 seconds), and corrects all the real-time flow monitoring data. Subsequently, this module uses the Ensemble Empirical Mode Decomposition (EEMD) technology to decompose the collected total building water consumption data signal into Intrinsic Mode Functions, and trains the water consumption time series data through the Long Short-Term Memory Network (LSTM) to capture the interdependence between data of each frequency and time, so as to accurately predict and export the building water consumption data, and finally form a big data sample of water usage habits, that is, a flow time curve table for 24 hours.

[0089] 3. Secondary water supply pipe network digital simulation system

[0090] The secondary water supply pipe network digital simulation system conducts further simulation and analysis based on the prediction results of the water usage information prediction system, specifically including a rapid modeling module for the topological structure of the water supply pipeline and a module for calculating the probability of pipeline head loss.

[0091] The rapid modeling module for the topological structure of the water supply pipeline quickly generates a topological structure model of the riser and the internal water supply pipeline of the building by importing the secondary water supply pipeline information of the case building collected by the historical water usage information collection module of the water usage information prediction system (including the water supply area, floor height, pipeline routing of risers and branch pipes, pipe materials, types and distributions of water using appliances, water supply pump stations, regulating water tanks and other water supply facilities).

[0092] The module for calculating the probability of pipeline head loss classifies and summarizes all the water using appliances inside the building based on the topological structure model of the internal water supply pipeline of the building, and randomly distributes the total available water consumption using the Poisson random distribution principle to form pipeline flow-pressure data. This module further calculates the head loss of the topological structure model of the internal water supply pipeline of the building under different flow conditions, and controls the head loss within a reasonable confidence interval. Finally, using the target head loss value, the constant pressure setting value at the user end, and the measured water tank liquid level value, the calculated value of the pump group outlet pressure is calculated.

[0093] In a preferred embodiment, the module for calculating the probability of pipeline head loss specifically includes: a random distribution sub-module, a confidence interval calculation sub-module, a target head loss value calculation sub-module, and a pump group outlet pressure calculation sub-module.

[0094] The random allocation sub-module is used to receive and process the classified summary information of the water-using appliances inside the building. Adopting the Poisson random allocation principle, it randomly allocates the water consumption of all the water-using appliances inside the building to generate a pipeline flow rate and pressure data set. Specifically, taking the maximum water consumption of each type of building stipulated in the relevant design standards as the maximum available water consumption for allocation, and the minimum water consumption appliance equivalent as the minimum available water consumption for allocation. And classify and summarize all the water-using appliances inside the building, and set them in ascending order of the total water volume. Since the water consumption of each water-using appliance is random, the Poisson random allocation principle is adopted. The total available water consumption is randomly allocated to the water consumption of the actual existing water-using appliances in the structure in a way that gradually increases with the increase of the water consumption equivalent of the water-using appliances in the building, that is, each time a ΔQ water volume is superimposed, and the water consumption of each water consumption is calculated and analyzed for tens of thousands of times of water-using appliance allocations. The total allocation calculation forms the pipeline flow rate - pressure data at a frequency of 0.3 - 0.5 m 3 / (h * ten thousand times), and initially obtains the head loss and the calculated value of the pump group outlet pressure of the building internal water supply pipeline topology model under different flow rates.

[0095] The confidence interval calculation sub-module calculates the maximum and minimum values of the head loss at different flow rates based on the data set of the random allocation sub-module and sets the confidence interval. Specifically, the pipeline head loss probability calculation module needs to control the head loss of the building internal water supply pipeline topology model within a reasonable confidence interval. As the water flow rate in the pipeline increases, the predicted range of the pipeline head loss becomes larger. Therefore, 20% and 95% of the maximum head loss of the building internal water supply pipeline topology model under different flow rates are used as the minimum and maximum head loss values that can be trusted.

[0096] The target head loss value calculation sub-module is used to calculate the target head loss value within the confidence interval. The specific implementation method is to calculate the average value of the minimum head loss and the maximum head loss, and use this average value as the target head loss value within this confidence interval.

[0097] The pump set outlet pressure calculation sub-module calculates the pump set outlet pressure value comprehensively by combining the target head loss value output by the target head loss value calculation sub-module, the preset constant pressure setting value of the user terminal, and the water tank liquid level value transmitted in real time by the Internet of Things perception system. The specific calculation process is as follows: According to the flow time curve obtained in real time by the water use information prediction system, for the predicted flow corresponding to a specific time point, first, the target head loss value calculation sub-module calculates the target head loss value at this flow rate. Subsequently, the target head loss value is added to the preset constant pressure setting value of the user terminal to obtain the outlet pressure value of each regional water distribution pipe. Finally, a subtraction operation is performed between the outlet pressure value of each regional water distribution pipe and the water tank liquid level value transmitted by the Internet of Things perception system to obtain the required pump head under this flow condition, and then the calculated value of the pump set outlet pressure is determined.

[0098] To ensure the accuracy of the model, in a preferred embodiment, the topological structure model of the building internal water supply pipeline established by the secondary water supply pipe network digital simulation system needs to be parameter-checked, aiming to make the pressure at the user terminal as stable as possible. By comparing the measured pressure value of the most unfavorable control point at the end of the pipe network on the highest floor monitored by the Internet of Things perception system with the end pressure value calculated by the model simulation, the pipe friction coefficient and local resistance coefficient are continuously corrected, the water supply pipeline model is checked, and the head loss relationship diagram is corrected, so that the error of the topological structure model of the building internal water supply pipeline is controlled within a certain range to ensure the correctness of the pipe network model.

[0099] Specifically, the specific adjustment method for parameter-checking of the topological structure model of the building internal water supply pipeline established by the secondary water supply pipe network digital simulation system is as follows: The pipe friction coefficient is adjusted within the range of Hazen-Williams coefficient 90 - 130, and at the same time, the local resistance coefficient is adjusted within the range of 0.1 - 10.0. The head loss and the calculated value of the pump set outlet pressure of the topological structure model of the building internal water supply pipeline are recalculated. When the error between the calculated value of the pump set outlet pressure and the measured value of the pump set outlet pressure is stable within 5% of the building floor height, the model parameter-checking is completed.

[0100] After long-term actual operation, it is also necessary to dynamically check the topological structure model of the building internal water supply pipeline according to the monitoring data of the Internet of Things perception system. When the error exceeds a certain value, it needs to be adjusted by 10%. Finally, the PLC frequency conversion controller is adjusted according to the calculated value of the pump set outlet pressure corresponding to a certain time.

[0101] 4. Pump set intelligent control system

[0102] The intelligent control system for the pump group sets a constant pressure value at the user end as the constant pressure setting value for the user end, and performs intelligent control based on the measured value of the pump group outlet pressure obtained by the Internet of Things perception system and the calculated value of the pump group outlet pressure obtained by the digital simulation system of the secondary water supply pipe network. The specific control method is as follows: The PLC frequency conversion controller is used to control and adjust the pump group frequency converter, so that the measured value of the pump group outlet pressure is maintained within the range of 100% - 120% of the calculated value of the pump group outlet pressure, to meet the water supply demand at different times and minimize the frequent change of the pressure of the end users.

[0103] 5. Computer Management and Cloud Platform System

[0104] The computer management and cloud platform system includes an edge computing module and a human-machine interaction interface.

[0105] The edge computing module is responsible for storing and calculating and analyzing the pipeline flow-pressure data obtained by the water consumption information prediction system and the digital simulation system of the secondary water supply pipe network, and also has functions such as abnormal data processing, pump group alarm, and data push. These data can be synchronously pushed to the cloud, client, and mobile terminal through remote communication for users to query and export at any time. The edge computing module can also analyze the energy consumption of the pump group by obtaining the data of time and pump group power consumption, that is, process the data of the instantaneous power consumption and cumulative power consumption of the pump group according to the operation situation of the pump group, and obtain the relationship curve between the operation time of the pump group and the cumulative power consumption.

[0106] The human-machine interaction interface displays the basic information of the community, the building pipeline model, the pump group performance curve, the real-time monitoring data, the pump group alarm setting, the system control setting, and the system control log, etc. for the user, providing an intuitive and convenient operation experience. The specific functions include the dynamic configuration management of the monitoring object, the dynamic binding of the communication protocol with the PLC, the abnormal alarm of the monitoring data, the dynamic update of the upload and download of the model, the curve query and analysis, the system operation log, the computing resource monitoring of the edge computing, etc.

[0107] Embodiment 2:

[0108] Based on the same design concept, this embodiment also provides a secondary water supply end user voltage stabilization and energy saving control method based on water consumption prediction. This control method is based on the control device including the Internet of Things perception system, water consumption information prediction system, digital simulation system of the secondary water supply pipe network, intelligent control system of the pump group, and computer management and cloud platform system provided in Embodiment 1.

[0109] Combined with Figure 1 As shown, the control method specifically includes the following steps:

[0110] Step S0, system construction.

[0111] Construct a control device including an Internet of Things perception system, a water consumption information prediction system, a digital simulation system for the secondary water supply pipe network, a pump group intelligent control system, and a computer management and cloud platform system. These systems cooperate with each other to jointly achieve the voltage stabilization and energy saving control of the end users of the secondary water supply based on the predicted water consumption.

[0112] Step S1, data monitoring and collection.

[0113] The Internet of Things perception system is used to monitor and collect in real time the data of the pump operation frequency, operation power consumption, pump group outlet flow, pump group outlet pressure, water tank level, and the pressure at the most unfavorable control point at the end of the pipe network on the highest floor. These data provide the basis for subsequent analysis and prediction.

[0114] Step S2, water consumption prediction.

[0115] The water consumption information prediction system is used to predict the water consumption based on historical water consumption data and real-time monitoring data. The specific steps are as follows:

[0116] Step S2.1, the flow sensor collects flow data in real time at a preset frequency (such as collecting one data every 6 seconds).

[0117] Step S2.2, data filtering technology is used to filter the collected flow data to improve the accuracy and reliability of the data.

[0118] Step S2.3, the filtered data is input into the Ensemble Empirical Mode Decomposition (EEMD) module to decompose the water consumption signal into Intrinsic Mode Functions.

[0119] Step S2.4, the decomposed Intrinsic Mode Functions are input into the Long Short-Term Memory network (LSTM) for training to capture the potential patterns of the water consumption time series and predict the water consumption time series for the next 24 hours.

[0120] Step S3, establishment of the water supply pipeline model and calculation of the head loss.

[0121] The digital simulation system for the secondary water supply pipe network establishes a topological structure model of the internal building water supply pipeline according to the predicted water consumption, and calculates the confidence interval of the head loss and the target head loss value under different flow rates. The specific steps are as follows:

[0122] Step S3.1, import the secondary water supply pipeline information of the case building collected by the historical water consumption information collection module in the water consumption information prediction system, including the water supply area, floor height, pipeline routing of the riser and branch pipes, pipe material type, types and distributions of water using appliances, etc.

[0123] Step S3.2, based on the imported information, automatically generate a topological structure model of the internal building water supply pipeline including risers, branch pipes, and water using appliance nodes.

[0124] Step S3.3: Using the Poisson random distribution principle, randomly distribute the total available water consumption to the water consumption of the actually existing water appliances in the structure in a way that the water equivalent increment of the water appliances in the building gradually increases, so as to obtain the head loss of the topological structure model of the internal building water supply pipeline under different flow rates.

[0125] Specifically, the pipeline head loss probability calculation module of the secondary water supply pipe network digital simulation system takes the maximum water consumption of each type of building specified in the relevant design standards as the maximum value of the available water consumption for distribution, and the minimum water appliance equivalent of the water consumption as the minimum value of the available water consumption for distribution. And classify and summarize all the water appliances inside the building, and set them in ascending order of the total water volume. Since the water consumption of each water appliance is random, the Poisson random distribution principle is adopted to randomly distribute the total available water consumption to the water consumption of the actually existing water appliances in the structure in a way that the water equivalent increment of the water appliances in the building gradually increases, that is, each time a ΔQ water volume is superimposed, and the water consumption of each water appliance is calculated and analyzed for tens of thousands of times. The total distribution calculation forms pipeline flow-pressure data according to the number of times of 0.3 - 0.5 m 3 / (h * ten thousand times), and initially obtains the head loss of the topological structure model of the internal building water supply pipeline under different flow rates.

[0126] Step S3.4: Set the head loss confidence interval, and take 20% of the maximum head loss value of the topological structure model of the internal building water supply pipeline under different flow rates as the minimum believable head loss value, and 95% as the maximum head loss value.

[0127] Step S3.5: Calculate the average value of the minimum head loss value and the maximum head loss value, and set this average value as the target head loss value.

[0128] Step S4: Calculate the pump group outlet pressure.

[0129] Add the target head loss value to the user end constant pressure setting value, and subtract the measured value of the water tank liquid level to obtain the calculated value of the pump group outlet pressure. The specific steps are as follows:

[0130] Step S4.1: According to the flow time curve obtained by the water use information prediction system, at the predicted flow rate corresponding to a certain time, the pipeline head loss probability calculation module of the secondary water supply pipe network digital simulation system obtains the target head loss value at this flow rate.

[0131] Step S4.2: Add the target head loss value at this flow rate to the user end constant pressure setting value to obtain the outlet pressure of each regional distribution pipe.

[0132] Step S4.3: Subtract the outlet pressure of each regional water distribution pipe from the water tank liquid level value transmitted by the Internet of Things perception system to obtain the required pump head at this flow rate, and then obtain the calculated value of the pump set outlet pressure.

[0133] Step S5: Model parameter calibration.

[0134] In a preferred embodiment, the topological structure model of the building internal water supply pipeline established by the secondary water supply network digital simulation system needs to be parameter-calibrated, aiming to make the pressure at the user end as stable as possible. According to the measured pressure value of the most unfavorable control point at the end of the pipe network on the highest floor monitored by the Internet of Things perception system, compare this value with the end pressure value calculated by the model simulation, continuously correct the pipe friction coefficient and local resistance coefficient, calibrate the water supply pipeline model, correct the head loss relationship diagram, and control the error of the topological structure model of the building internal water supply pipeline within a certain range to ensure the correctness of the pipe network model. The specific steps are as follows:

[0135] Step S5.1: Compare the measured pressure value of the most unfavorable control point at the end of the pipe network on the highest floor by the Internet of Things perception system with the end pressure value calculated by the topological structure model of the building internal water supply pipeline.

[0136] Step S5.2: Dynamically adjust the pipe friction coefficient and local resistance coefficient according to the comparison result. The specific adjustment method is to adjust the pipe friction coefficient within the range of Hazen-Williams coefficient from 90 to 130, and at the same time adjust the local resistance coefficient within the range of 0.1 to 10.0. And recalculate the head loss and pump set outlet pressure of the topological structure model of the building internal water supply pipeline.

[0137] Step S5.3: Repeat the above steps until the error between the calculated value of the pump set outlet pressure and the measured value of the pump set outlet pressure is stable within 5% of the floor height, then the model parameter calibration is completed.

[0138] In addition, after actual long-term operation, it is also necessary to dynamically calibrate the topological structure model of the building internal water supply pipeline according to the monitoring data of the Internet of Things perception system. When the error exceeds a certain value, it needs to be adjusted by 10%. Finally, adjust the PLC frequency conversion controller according to the calculated value of the pump set outlet pressure corresponding to a certain time.

[0139] Step S6: Intelligent control of the pump set.

[0140] The intelligent control system of the pump set dynamically adjusts the operating frequency of the pump set through the PLC frequency conversion controller, so that the measured value of the pump set outlet pressure is stable within the range of 100% - 120% of the calculated value of the pump set outlet pressure, to meet the water supply demand at different times and minimize the frequent change of the pressure at the end user.

[0141] Step S7: Energy consumption analysis of the pump set.

[0142] In a preferred embodiment, the edge computing module of the computer management and cloud platform system can analyze the energy consumption of the pump group. The specific steps are as follows:

[0143] Step S7.1: Receive and record the time data of the pump group operation and the corresponding pump group power consumption data.

[0144] Step S7.2: Process the instantaneous power consumption data and cumulative power consumption data of the pump group based on the actual operation of the pump group.

[0145] Step S7.3: Generate and output the relationship curve between the pump group operation time and the cumulative power consumption according to the processing results, providing data support for the energy efficiency evaluation and optimization of the pump group.

[0146] In summary, for the secondary water supply end-user voltage stabilization and energy-saving control method provided by the present invention, the flow sensor of the Internet of Things perception system transmits flow data to the water usage information prediction system. After the water usage information prediction system corrects the flow data, it uses intelligent algorithms to train the data and predicts the water demand of the building. Based on the water demand prediction of the water usage information prediction system, the secondary water supply pipe network digital simulation system collects information on the building pipeline and appliance structures and analyzes the single water usage amount to obtain the head loss of the pipeline topology model corresponding to different flows at different times in the building. After calculating the confidence interval range of the possible head loss under different flow conditions through the pipeline head loss probability calculation module, the target head loss value is obtained. After adding the user end constant pressure setting value set by the pump group intelligent control system and subtracting the actual water tank liquid level value detected by the water tank liquid level gauge of the Internet of Things perception system, the required pump head at this flow is obtained, and then the pump outlet pressure calculation value is obtained. The pump outlet pressure calculation value is checked by the measured end pressure value and the measured pump outlet pressure value obtained by the Internet of Things perception system. Based on the pump outlet pressure calculation value obtained by the secondary water supply pipe network digital simulation system, the pump group intelligent control system uses the PLC controller to adjust the pump group frequency converter so that the measured pump group outlet pressure value reaches the range of 100% - 120% of the pump outlet pressure calculation value. The computer management and cloud platform system can store the data transmitted by the Internet of Things perception system, the secondary water supply pipe network digital simulation system, and the pump group intelligent control system through remote communication, alarm for abnormal situations, and use the edge computing core module to process the data and conduct power consumption statistics. The human-computer interaction platform of the computer management and cloud platform system can push the data to the client, cloud, and mobile terminals, and users can query and export the data on different platforms and view the system operation logs. On the premise of real-time prediction of water usage in different buildings, the present invention obtains the pipeline head loss corresponding to different flows at different times of the day and the pump group outlet pressure that should be satisfied when the end-user pressure is constant. Only by installing a pressure sensor at the most unfavorable control point at the end of the pipe network on the highest floor can the frequency of the pump group intelligent control system be dynamically adjusted through the system to meet the water supply requirements of different buildings at different times and minimize the frequent changes in the end-user pressure. At the same time, the computer management system and cloud platform can be used to analyze and remotely adjust the data, improving the existing ordinary variable frequency constant pressure water supply technology.

[0147] Implementation case:

[0148] A commercial office project in Shanghai. This project has a 14-story commercial office building with a total construction area of 70,000 m 2 . The basement and the first floor of this building are directly supplied by the municipal water supply, and the 2nd - 14th floors are supplied by the pump group and the water tank in the domestic pump room. The designed maximum daily water consumption is 225 m3 / d, with the maximum daily water consumption of 22.3 m 3 / d. A constant-pressure variable-frequency water supply pump group at the pump outlet is adopted for water supply. There are 3 pumps in a group, which are used as spares for each other. The set pressure at the constant pressure outlet of the pump group is 1.10 Mpa.

[0149] Now, a secondary water supply end-user voltage stabilization and energy-saving control transformation based on real-time water consumption prediction is carried out for this project. Through the transformation of the system to constant-pressure water supply at the end, after the transformation, it is variable-pressure variable-frequency water supply at the outlet of the end constant-pressure pump. After the transformation, the end constant pressure at the most unfavorable point is 0.2 - 0.3 Mpa. Before the transformation, the total water supply volume and power consumption for 4 consecutive days in the starting constant-pressure state were 6.90 m 3 / d and 39.10 kW*h( Figure 2 、 Figure 3 ). After the transformation, the total water supply volume and power consumption for 4 consecutive days in the end constant-pressure state were 7.17 m 3 / d and 28.70 kW*h( Figure 2 、 Figure 3 ). This means that while the water supply volume is increased, the system energy consumption is greatly reduced. Further calculations show that compared with before the transformation, the average energy consumption per thousand tons of water is saved by 31.79%( Figure 4 ).

[0150] The specific embodiments of the present invention have been described in detail above, but they are only examples, and the present invention is not limited to the specific embodiments described above. For those skilled in the art, any equivalent modifications and substitutions made to the present invention are also within the scope of the present invention. Therefore, equivalent transformations and modifications made without departing from the spirit and scope of the present invention should be covered within the scope of the present invention.

Claims

1. A voltage stabilizing and energy-saving control device for secondary water supply end users based on water consumption prediction, characterized in that: include: The IoT sensing system monitors and collects data on the operating frequency, power consumption, pump outlet flow, pump outlet pressure, water tank level, and the most unfavorable control point pressure at the end of the highest floor pipe network in real time; The water consumption information prediction system, based on the data collected by the IoT sensing system, collects building information through the historical water consumption information collection module and establishes a secondary water supply pipeline layout template, and predicts the water consumption of the building through the water consumption time series prediction module; Second, the digital simulation system of the water supply network generates the topological structure model of the water supply pipeline inside the building through the water supply pipeline topological structure rapid modeling module according to the prediction results of the water use information prediction system, and calculates the head loss and pump group outlet pressure calculation value under different flow rates through the pipeline head loss probability calculation module; The intelligent control system of the pump group dynamically adjusts the operating frequency of the pump group through the PLC frequency conversion controller according to the calculated value of the pump group outlet pressure calculated by the digital simulation system of the secondary supply network, so that the measured value of the pump group outlet pressure is stable within the range of 100% to 120% of the calculated value of the pump group outlet pressure; The computer management and cloud platform system includes an edge computing module and a human-computer interaction interface. The edge computing module is used to store, calculate and analyze, handle exceptions and remotely push data obtained from the water use information prediction system and the secondary supply network digital simulation system. The human-computer interaction interface is used to display system information and control settings.

2. The device according to claim 1, characterized in that The historical water use information collection module of the water use information prediction system includes: The basic information collection submodule is used to collect basic information including water supply area, floor area, floor height, and pipeline direction; The key information collection submodule is used to collect key information of the independent water supply system including the water tank, the electric control device, and the pressure water container; The secondary water supply pipeline layout template library construction submodule is used to establish the secondary water supply pipeline layout template library for different types of building floors based on the collected basic information and key information, and provide tools for editing floor templates to support users to dynamically edit and supplement templates based on floor height and water-using appliance distribution; The water usage database construction submodule is used to design water usage orders based on the building water supply and drainage appliances, and to build a water usage database for different water-using appliances based on the actual seasons, north-south regional differences, and historical data on water usage of actual water-using appliances.

3. The device according to claim 1, characterized in that The water consumption time series prediction module of the water consumption information prediction system includes: a data filtering submodule, an enhanced empirical mode decomposition submodule and a long short-term memory network submodule; The data filtering submodule is used to receive the real-time monitoring data in the IoT sensing system, perform filtering processing using data filtering technology, and correct all real-time traffic monitoring data; The enhanced empirical mode decomposition submodule is used to perform intrinsic mode function decomposition on the flow data signal processed by the data filtering submodule, and decompose the flow data signal into intrinsic mode functions using enhanced empirical mode decomposition EEMD technology; The long short-term memory network submodule is used to train the water consumption time series data decomposed by the enhanced empirical mode decomposition submodule, and use the deep learning algorithm of the long short-term memory network LSTM to capture the interdependence between each frequency data and time, predict and export the building water consumption data, and form a big data sample of water use habits, that is, a 24-hour flow time curve table.

4. The device according to claim 1, characterized in that The water supply pipeline topology structure rapid modeling module of the secondary supply pipe network digital simulation system includes: a case information import submodule and a topology structure generation submodule; The case information import submodule is used to receive and import the case building secondary water supply pipeline information collected by the historical water use information collection module in the water use information prediction system; The topology structure generation submodule is used to quickly generate a topology structure model of pipes and water supply pipelines inside a building based on the detailed information imported by the case information import submodule.

5. The device according to claim 1, characterized in that The pipeline head loss probability calculation module of the secondary supply network digital simulation system includes: a random allocation submodule, a confidence interval calculation submodule, a target head loss value calculation submodule and a pump group outlet pressure calculation submodule; The random allocation submodule is used to receive and process the classified summary information of water-using appliances in the building, and randomly allocate the water consumption of all water-using appliances in the building using the Poisson random allocation principle to generate a pipeline flow and pressure data set; The confidence interval calculation submodule calculates the maximum and minimum values ​​of the head loss under different flow rates based on the data set of the random allocation submodule, and sets the confidence interval; The target water head loss value calculation submodule is used to calculate the target water head loss value within the confidence interval; The pump group outlet pressure calculation submodule combines the target head loss value output by the target head loss value calculation submodule, the preset user terminal constant pressure setting value and the water tank liquid level value transmitted in real time by the Internet of Things sensing system to comprehensively calculate the pump group outlet pressure value. The specific calculation process is: adding the target head loss value to the user terminal constant pressure setting value, and then deducting the actual measured value of the water tank liquid level from this sum to obtain the calculated value of the pump group outlet pressure.

6. A voltage stabilization and energy-saving control method for secondary water supply end users based on water consumption prediction, characterized in that: Based on the voltage stabilizing and energy-saving control device for secondary water supply end users based on water consumption prediction as described in any one of claims 1 to 5, the control method comprises the following steps: S1, through the IoT sensing system, real-time monitoring and collection of data on water pump operating frequency, operating power consumption, pump group outlet flow, pump group outlet pressure, water tank liquid level, and the most unfavorable control point pressure at the end of the highest floor pipe network; S2, using the water consumption information prediction system, based on historical water consumption data and real-time monitoring data, using enhanced empirical mode decomposition (EEMD) technology to decompose the water consumption signal into intrinsic mode functions, and using long short-term memory network (LSTM) to train time series data to predict the water consumption of buildings; S3, the secondary supply network digital simulation system establishes the topological structure model of the water supply pipeline inside the building according to the predicted water consumption, simulates the random distribution of water consumption based on the Poisson random distribution principle, calculates the confidence interval of the head loss under different flow rates, and determines the target head loss value; S4, adding the target head loss value to the preset user terminal constant pressure setting value, and deducting the measured value of the water tank liquid level to obtain the calculated value of the pump group outlet pressure; S5, the intelligent control system of the pump group dynamically adjusts the operating frequency of the pump group through the PLC frequency conversion controller, so that the measured value of the pump group outlet pressure is stabilized in the range of 100% to 120% of the calculated value of the pump group outlet pressure.

7. The method according to claim 6, characterized in that Step S2 specifically includes the following steps: The flow sensor collects flow data in real time at a preset frequency; Use data filtering technology to filter the collected traffic data; The data corrected by filtering technology is input into the EEMD submodule; The decomposed intrinsic mode function is input into the LSTM network for training, and the time series of water consumption in the next 24 hours is output.

8. The method according to claim 6, characterized in that In step S3, the method for establishing the topological structure model of the water supply pipeline inside the building includes the following steps: Importing the secondary water supply pipeline information of the case building collected by the historical water use information collection module in the water use information prediction system; Based on the imported secondary water supply pipeline information of the case building, the topological structure model of the building's internal water supply pipeline, including risers, branches and water-using appliance nodes, is automatically generated.

9. The method according to claim 6, characterized in that In step S3, the calculation steps of the head loss confidence interval and the target head loss value include: The pipeline head loss probability calculation module sets the maximum water consumption of each type of building as the upper limit of the water consumption that can be allocated according to the relevant design standards, and the water consumption of the smallest water-using appliance as the lower limit of the water consumption that can be allocated; All water-using appliances in the building are classified and summarized, and sorted from small to large according to the total water volume. The Poisson random allocation principle is used to randomly allocate the total water volume available for allocation in a way that the water consumption of the water-using appliances in the building gradually increases. The water head loss of the topological structure model of the water supply pipeline inside the building under different flow rates is obtained; the total allocation calculation times of the Poisson random allocation are 0.3-0.5m 3 / (h·10,000 times), each time the water volume ΔQ is allocated and superimposed to generate a pipeline flow-pressure data set; The 20% of the maximum head loss of the topological structure model of the water supply pipeline inside the building under different flow rates is taken as the credible minimum head loss value, and 95% is taken as the credible maximum head loss value; The average value of the minimum head loss value and the maximum head loss value is calculated, and the average value is set as the target head loss value.

10. The method according to claim 6, characterized in that Step S4 specifically includes the following steps: According to the flow time curve obtained by the water use information prediction system, under the predicted flow corresponding to a certain time, the pipeline head loss probability calculation module of the secondary supply network digital simulation system obtains the target head loss value under the flow; The target head loss value under the flow rate is added to the preset user terminal constant pressure setting value to obtain the outlet pressure of each regional water distribution pipe; The outlet pressure of each regional water distribution pipe is subtracted from the water tank liquid level value transmitted by the Internet of Things sensing system to obtain the required water pump head under the flow rate, and then the calculated value of the pump group outlet pressure is obtained.

11. The method according to claim 6, characterized in that The method further comprises: performing parameter verification on the topological structure model of the water supply pipeline inside the building established by the digital simulation system of the second supply pipe network, specifically comprising the following steps: a) Compare the pressure value of the most unfavorable control point at the end of the highest floor pipe network measured by the IoT sensing system with the terminal pressure value simulated and calculated by the topological structure model of the water supply pipeline inside the building; b) According to the comparison results, dynamically adjust the pipeline friction coefficient and local resistance coefficient, and recalculate the head loss and pump group outlet pressure of the topological structure model of the water supply pipeline inside the building; c) Repeat a) and b) until the error between the calculated value of the pump group outlet pressure and the measured value of the pump group outlet pressure is stabilized within 5% of the floor height.

12. The method according to claim 11, characterized in that The method further includes: the dynamic adjustment of the pipeline friction coefficient and the local resistance coefficient includes: adjusting the pipeline friction coefficient within the range of Hessen-William coefficient 90 to 130, and adjusting the local resistance coefficient within the range of 0.1 to 10.0.

Images