Rural water supply pipe network pressure monitoring analysis method and device based on digital twinning

Abstract

The invention discloses a rural water supply pipe network pressure monitoring analysis method and device based on digital twinning. The method comprises the following steps: collecting historical basic data of a rural prediction area; flow and pressure sensors are arranged at nodes of a rural water supply network to obtain monitoring data; uploading the data to a server through a data stream processing framework, and preprocessing the data to obtain preprocessed data; constructing a relational database management system to perform batch processing on the preprocessed data to obtain historical basic data and monitoring data after batch processing; constructing a rural pipe network water pressure prediction model; constructing a pipe explosion detection and water pressure control model; according to the physical data, constructing a rural water supply pipe network geometric model based on ArcGIS; and constructing a rural water supply pipe network system data model and realizing visualization of the operation state of the water supply pipe network. Rural water supply characteristics and water consumption characteristics are fully considered, intelligent management and control of the rural water supply system are achieved, and reliability and safety of rural water supply are effectively guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of rural water supply, and in particular to a method and device for monitoring and analyzing the pressure of a rural water supply network based on digital twins. Background Art

[0002] Rural water supply is closely linked to people's daily lives and production. Rural water supply in the hilly southern region is characterized by a branched network, scattered water points, and dramatic elevation fluctuations along the pipelines. This results in significant variations in water pressure across different areas, compromising water supply reliability and network security.

[0003] At present, the water supply network in most rural areas still adopts traditional manual management methods, which cannot realize dynamic monitoring and control of water pressure in the water supply network pipelines. It has shortcomings such as low water supply guarantee efficiency, untimely response, and serious pipeline leakage, which has an adverse impact on the quality of rural water supply.

[0004] Therefore, it is necessary to develop a rural water supply network pressure monitoring and regulation method based on digital twins to realize intelligent management and control of water pressure in rural water supply networks. Summary of the Invention

[0005] To this end, the present invention provides a rural water supply network pressure monitoring and analysis method and device based on digital twins, which fully considers the characteristics of rural water supply and water use, realizes the intelligent management and control of the rural water supply system, and effectively ensures the reliability and safety of rural water supply.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a rural water supply network pressure monitoring and analysis method based on digital twins, comprising:

[0007] Collect the setting data of the rural forecast area and obtain the historical basic data of the rural forecast area;

[0008] Selecting nodes of the rural water supply network according to a node selection principle; deploying flow and pressure sensors at the selected nodes, and acquiring monitoring data through the flow and pressure sensors;

[0009] Uploading the historical basic data and the monitoring data to a server through a data stream processing framework; preprocessing the historical basic data and the monitoring data through the server to obtain preprocessed data;

[0010] Building a relational database management system based on MySQL; using Hadoop and Spark tools to batch process the pre-processed data through the relational database management system to obtain batch-processed historical basic data and monitoring data;

[0011] Constructing a rural pipe network water pressure prediction model; training the rural pipe network water pressure prediction model using the historical basic data to obtain the trained rural pipe network water pressure prediction model; inputting the batch-processed historical basic data and monitoring data into the trained rural pipe network water pressure prediction model to obtain rural pipe network water pressure prediction data through prediction;

[0012] Constructing a pipe burst detection and water pressure control model; inputting batch-processed monitoring data and the rural pipe network water pressure prediction data into the pipe burst detection and water pressure control model; through the detection and processing of the pipe burst detection and water pressure control model, outputting a list of high-risk pipe sections and performing PID control;

[0013] Collecting physical data of the rural water supply network; constructing a rural water supply network geometric model based on ArcGIS based on the physical data;

[0014] Constructing a rural water supply network system data model based on the rural pipe network water pressure prediction model, the pipe burst detection and water pressure control model, and the rural water supply network geometric model;

[0015] The back-end program of the rural water supply network system data model is written in Python language. The rural water supply network system data model outputs time series curves of real-time data and predicted data to realize the visualization of the operating status of the water supply network.

[0016] As a preferred solution for the rural water supply network pressure monitoring and analysis method based on digital twins, in the process of collecting the set data of the rural prediction area, the set data includes: historical water supply and actual water consumption hourly data, temperature and humidity data of corresponding time nodes, data on the day of the week corresponding to the time of the day, and numerical data on whether the day is a holiday.

[0017] As a preferred solution for the rural water supply network pressure monitoring and analysis method based on digital twins, in the process of selecting nodes of the rural water supply network according to the node selection principle, the node selection principle includes: laying at user entry points, network intersections, and at intervals on straight sections of the network.

[0018] As a preferred solution of the rural water supply network pressure monitoring and analysis method based on digital twins, in the process of preprocessing the historical basic data and the monitoring data by the server, for the historical basic data, the hourly data and the temperature and humidity data of the corresponding time node, holiday information, and day of the week data are stored as a data sequence;

[0019] For the monitoring data, the data of the hour and the predicted data obtained by the rural pipe network water pressure prediction model are stored as a data sequence. If there is no predicted data, the default value None is taken;

[0020] For abnormal data in the monitoring data, if the abnormal situation is the missing of some parameter values, the average of the corresponding data of the six hours before and after the hour is taken for filling; if the abnormal situation is the long-term repetition of parameter values, the data of the same time period of the same month is taken, multiplied by the deviation coefficient and used for filling, and the deviation coefficient is taken as the divisor average of the non-repeating data of the same time period of the same month and the non-repeating data of the repeated time period; if the abnormal situation is the parameter value fluctuation error, the average value of the data of the same time period in recent days is taken as filling.

[0021] As a preferred solution for the rural water supply network pressure monitoring and analysis method based on digital twins, the expression of the rural pipe network water pressure prediction model is:

[0022] pre.=αTime+βSpace+γH+σQ+ε

[0023] Where pre. is the predicted value of a node; Time is the temporal inertia term of the node, including the average water pressure of the node in the previous time period, the average water pressure of the same month and week, the average water pressure of the same year and quarter, and the average water pressure of the same quarter and week in the previous three years; Space is the spatial inertia term of the node, including the monitored water pressure of the upstream node in the same time period and the predicted water pressure of the upstream node in the same time period (this term does not exist for the initial calculation node); H is the benchmark term of the node, that is, the actual monitored water pressure of the node at the same time; Q is the water pressure correction term of the node, including the water volume monitoring values ​​of all nodes at the same time; ε is the bias term of the node, which is used to explain possible unconsidered rules in model learning; α, β, γ, and δ are the learning rates of the LSTM model, which are calibrated and corrected by the hyperparameter total lr and model learning.

[0024] As a preferred solution for the rural water supply network pressure monitoring and analysis method based on digital twins, the detection and processing steps of the pipe burst detection and water pressure control model are as follows:

[0025] The batch-processed monitoring data and the rural pipe network water pressure prediction data are identified by a decision tree model, the pipe burst risk is output, and the pipe burst risk is marked;

[0026] Optimize low-burst pipe risk data through fuzzy control strategy;

[0027] Time monitoring and processing of high-explosive pipe risk data; if the set time is exceeded, the log is automatically uploaded and the monitoring personnel are notified through the set method;

[0028] Analyze abnormal water pressure nodes and the possibility of pipe burst, output a list of high-risk pipe sections and perform PID control.

[0029] The present invention also provides a rural water supply network pressure monitoring and analysis device based on digital twins, based on the above rural water supply network pressure monitoring and analysis method based on digital twins, including:

[0030] A historical basic data collection module is used to collect the set data of the rural prediction area and obtain the historical basic data of the rural prediction area;

[0031] A monitoring data acquisition module is used to select nodes of the rural water supply network according to the node selection principle; deploy flow and pressure sensors at the selected nodes and acquire monitoring data through the flow and pressure sensors;

[0032] A data preprocessing module is used to upload the historical basic data and the monitoring data to a server through a data stream processing framework; and preprocess the historical basic data and the monitoring data through the server to obtain preprocessed data;

[0033] A data batch processing module is used to build a relational database management system based on MySQL; through the relational database management system, the pre-processed data is batch-processed using Hadoop and Spark tools to obtain batch-processed historical basic data and monitoring data;

[0034] A rural pipe network water pressure prediction model construction and processing module is used to construct a rural pipe network water pressure prediction model; train the rural pipe network water pressure prediction model using the historical basic data to obtain the trained rural pipe network water pressure prediction model; input the batch-processed historical basic data and monitoring data into the trained rural pipe network water pressure prediction model to obtain rural pipe network water pressure prediction data through prediction;

[0035] A pipe burst detection and water pressure control model construction and processing module is used to construct a pipe burst detection and water pressure control model; batch-processed monitoring data and the rural pipe network water pressure prediction data are input into the pipe burst detection and water pressure control model; through the pipe burst detection and water pressure control model detection and processing, a list of high-risk pipe sections is output and PID control is performed;

[0036] A rural water supply network geometric model building module is used to collect physical data of the rural water supply network; based on the physical data, a rural water supply network geometric model is built based on ArcGIS;

[0037] A rural water supply network system data model construction module, configured to construct a rural water supply network system data model based on the rural water supply network water pressure prediction model, the pipe burst detection and water pressure control model, and the rural water supply network geometric model;

[0038] The rural water supply network system data model visualization output module is used to write the back-end program of the rural water supply network system data model in Python language. The rural water supply network system data model outputs time series curves of real-time data and predicted data to realize the visualization of the operation status of the water supply network.

[0039] As a preferred solution for the rural water supply network pressure monitoring and analysis device based on digital twins, in the historical basic data collection module, in the process of collecting the set data of the rural prediction area, the set data includes: historical water supply and actual water use hour data, temperature and humidity data of corresponding time nodes, data on the day of the week corresponding to the time of the day, and numerical data on whether the day is a holiday.

[0040] As a preferred solution for the rural water supply network pressure monitoring and analysis device based on digital twins, in the monitoring data acquisition module, in the process of selecting nodes of the rural water supply network according to the node selection principle, the node selection principle includes: laying at user entry points, network intersections, and at intervals on straight sections of the network.

[0041] As a preferred solution of the rural water supply network pressure monitoring and analysis device based on digital twins, in the data preprocessing module, during the preprocessing of the historical basic data and the monitoring data by the server, the historical basic data is stored as a data sequence, along with the temperature and humidity data of the corresponding time node, holiday information, and day of the week data.

[0042] For the monitoring data, the data of the hour and the predicted data obtained by the rural pipe network water pressure prediction model are stored as a data sequence. If there is no predicted data, the default value None is taken;

[0043] For abnormal data in the monitoring data, if the abnormal situation is the missing of some parameter values, the average of the corresponding data of the six hours before and after the hour is taken for filling; if the abnormal situation is the long-term repetition of parameter values, the data of the same time period of the same month is taken, multiplied by the deviation coefficient and used for filling, and the deviation coefficient is taken as the divisor average of the non-repeating data of the same time period of the same month and the non-repeating data of the repeated time period; if the abnormal situation is the parameter value fluctuation error, the average value of the data of the same time period in recent days is taken as filling.

[0044] As a preferred solution of the rural water supply network pressure monitoring and analysis device based on digital twin, in the rural pipe network water pressure prediction model construction and processing module, the expression of the rural pipe network water pressure prediction model is:

[0045] pre.=αTime+βSpace+γH+σQ+ε

[0046] Where pre. is the predicted value of a node; Time is the temporal inertia term of the node, including the average water pressure of the node in the previous time period, the average water pressure of the same month and week, the average water pressure of the same year and quarter, and the average water pressure of the same quarter and week in the previous three years; Space is the spatial inertia term of the node, including the monitored water pressure of the upstream node in the same time period and the predicted water pressure of the upstream node in the same time period (this term does not exist for the initial calculation node); H is the benchmark term of the node, that is, the actual monitored water pressure of the node at the same time; Q is the water pressure correction term of the node, including the water volume monitoring values ​​of all nodes at the same time; ε is the bias term of the node, which is used to explain possible unconsidered rules in model learning; α, β, γ, and δ are the learning rates of the LSTM model, which are calibrated and corrected by the hyperparameter total lr and model learning.

[0047] As a preferred solution for the rural water supply network pressure monitoring and analysis device based on digital twins, the detection and evaluation submodules in the pipe burst detection and water pressure control model construction and processing module include:

[0048] A data identification and marking submodule is used to identify the batch-processed monitoring data and the rural pipe network water pressure prediction data through a decision tree model, output the pipe burst risk, and mark the pipe burst risk;

[0049] Low-explosion pipe risk data optimization submodule, used to optimize low-explosion pipe risk data through fuzzy control strategy;

[0050] The high-explosive pipe risk data processing submodule is used to monitor and process high-explosive pipe risk data over time. If the set time is exceeded, the log is automatically uploaded and the monitoring personnel are notified through the set method.

[0051] The possibility analysis submodule is used to analyze abnormal water pressure nodes and the possibility of pipe burst, output a list of high-risk pipe sections and perform PID control.

[0052] The present invention has the following advantages: collecting set data of rural prediction areas to obtain historical basic data of rural prediction areas; selecting nodes of rural water supply pipe networks according to node selection principles; deploying flow and pressure sensors at selected nodes to obtain monitoring data through the flow and pressure sensors; uploading the historical basic data and the monitoring data to a server through a data stream processing framework; preprocessing the historical basic data and the monitoring data through the server to obtain preprocessed data; building a relational database management system based on MySQL; batch processing the preprocessed data using Hadoop and Spark tools through the relational database management system to obtain batch-processed historical basic data and monitoring data; building a rural pipe network water pressure prediction model; training the rural pipe network water pressure prediction model through the historical basic data to obtain the trained rural pipe network water pressure prediction model; and The basic data and monitoring data are input into the trained rural pipe network water pressure prediction model, and the rural pipe network water pressure prediction data is obtained through prediction; a pipe burst detection and water pressure control model is constructed; the batch-processed monitoring data and the rural pipe network water pressure prediction data are input into the pipe burst detection and water pressure control model, and the pipe burst detection and water pressure control model is used for detection and processing, and a list of high-risk pipe sections is output and PID control is performed; the physical data of the rural water supply network is collected; based on the physical data, a rural water supply network geometric model is constructed based on ArcGIS; based on the rural pipe network water pressure prediction model, the pipe burst detection and water pressure control model and the rural water supply network geometric model, a rural water supply network system data model is constructed; the back-end program of the rural water supply network system data model is written in Python language, and the rural water supply network system data model outputs time series curves of real-time data and prediction data to realize visualization of the operation status of the water supply network. This invention analyzes historical data, mines data features, and constructs a water pressure prediction model based on digital twins. This allows for precise monitoring and early warning of water pressure across the entire water supply network, ensuring water supply safety. This invention fully considers the characteristics of rural water supply and water usage, enabling intelligent management and control of rural water supply systems, effectively ensuring the reliability and safety of rural water supply. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are merely exemplary, and those skilled in the art can, without inventive effort, derive other implementation drawings based on the provided drawings.

[0054] The structures, proportions, sizes, etc. illustrated in this specification are intended solely to complement the contents disclosed herein and to facilitate understanding and reading by persons skilled in the art. They are not intended to limit the conditions under which the present invention may be implemented and therefore have no substantive technical significance. Any structural modifications, changes in proportions, or adjustments in sizes, without affecting the efficacy and objectives of the present invention, shall remain within the scope of the technical contents disclosed herein.

[0055] Figure 1 This is a flow chart of the rural water supply network pressure monitoring and analysis method based on digital twins provided in Example 1 of the present invention;

[0056] Figure 2 This is a schematic diagram of a specific implementation process of the rural water supply network pressure monitoring and analysis method based on digital twins provided in Example 1 of the present invention;

[0057] Figure 3 This is a schematic diagram of the implementation process of the burst pipe detection and water pressure control model in the rural water supply network pressure monitoring and analysis method based on digital twins provided in Example 1 of the present invention;

[0058] Figure 4 This is a schematic diagram of the anomaly detection probability in the rural water supply network pressure monitoring and analysis method based on digital twins provided in Example 1 of the present invention;

[0059] Figure 5 This is a schematic diagram of monitoring water pressure data in the rural water supply network pressure monitoring and analysis method based on digital twins provided in Example 1 of the present invention;

[0060] Figure 6 This is a schematic diagram of predicted water pressure data in the rural water supply network pressure monitoring and analysis method based on digital twins provided in Example 1 of the present invention;

[0061] Figure 7 This is a schematic diagram comparing water pressure monitoring data and predicted data in the rural water supply network pressure monitoring and analysis method based on digital twins provided in Example 1 of the present invention;

[0062] Figure 8 This is a schematic diagram of the fuzzy control process in the rural water supply network pressure monitoring and analysis method based on digital twins provided in Example 1 of the present invention;

[0063] Figure 9 Schematic diagram of error accumulation elimination in the rural water supply network pressure monitoring and analysis method based on digital twin provided in Example 1 of the present invention;

[0064] Figure 10 This is a schematic diagram of the architecture of the rural water supply network pressure monitoring and analysis device based on digital twin provided in Example 2 of the present invention. DETAILED DESCRIPTION

[0065] The following describes the implementation of the present invention using specific embodiments. Those skilled in the art will readily understand the other advantages and benefits of the present invention from the disclosure herein. Obviously, the embodiments described are only a portion of the present invention, not all of it. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without inventive effort are intended to fall within the scope of protection of the present invention.

[0066] Example 1

[0067] See also Figure 1 and Figure 2 , embodiment 1 of the present invention provides a rural water supply network pressure monitoring and analysis method based on digital twin, comprising the following steps:

[0068] S1. Collect the setting data of the rural prediction area and obtain the historical basic data of the rural prediction area;

[0069] S2. Select nodes of the rural water supply network according to the node selection principle; deploy flow and pressure sensors at the selected nodes, and obtain monitoring data through the flow and pressure sensors;

[0070] S3. Uploading the historical basic data and the monitoring data to a server through a data stream processing framework; preprocessing the historical basic data and the monitoring data through the server to obtain preprocessed data;

[0071] S4. Building a relational database management system based on MySQL; using Hadoop and Spark tools to batch process the pre-processed data through the relational database management system to obtain batch-processed historical basic data and monitoring data;

[0072] S5. Construct a rural pipe network water pressure prediction model; train the rural pipe network water pressure prediction model using the historical basic data to obtain the trained rural pipe network water pressure prediction model; input the batch-processed historical basic data and monitoring data into the trained rural pipe network water pressure prediction model to obtain rural pipe network water pressure prediction data through prediction;

[0073] S6. Constructing a pipe burst detection and water pressure control model; inputting the batch-processed monitoring data and the rural pipe network water pressure prediction data into the pipe burst detection and water pressure control model; outputting a list of high-risk pipe sections through detection and processing by the pipe burst detection and water pressure control model and performing PID control;

[0074] S7. Collecting physical data of the rural water supply network; and constructing a rural water supply network geometric model based on ArcGIS according to the physical data;

[0075] S8. Constructing a rural water supply network system data model based on the rural pipe network water pressure prediction model, the pipe burst detection and water pressure control model, and the rural water supply network geometric model;

[0076] S9. Use Python language to write the back-end program of the rural water supply network system data model, and the rural water supply network system data model outputs time series curves of real-time data and predicted data to realize the visualization of the operation status of the water supply network.

[0077] In this embodiment, in step S1, the set data of the rural prediction area is collected to obtain the historical basic data of the rural prediction area;

[0078] The setting data includes: historical water supply and actual water consumption hourly data, temperature and humidity data at corresponding time nodes, data on the day of the week corresponding to the time of the day, and numerical data on whether the day is a holiday.

[0079] Specifically, the time corresponding to the day is the day of the week. Assuming that the date data format is the Gregorian calendar date, the calculation is performed using the Zeiller formula, and the calculation formula is:

[0080]

[0081] Where h is the result of the day of the week, 0 represents Saturday, 1 represents Sunday, 2 represents Monday, and so on; q is the day of the date (1 to 31); m is the month (3 represents March, 4 represents April, ..., 12 represents December; 1 and 2 represent the 13th and 14th months of the previous year, that is, January and February); K is the last two digits of the year; J is the first two digits of the year.

[0082] Whether the current day is a holiday is numerically represented by the data. The mapping rule is as shown in the following function H(x):

[0083]

[0084] In this embodiment, in step S2, nodes of the rural water supply network are selected according to the node selection principle; flow and pressure sensors are deployed at the selected nodes, and monitoring data is obtained through the flow and pressure sensors;

[0085] The node selection principles include: laying at user entry points, pipe network intersections, and at intervals on straight sections of the pipe network.

[0086] Specifically, important nodes of the rural water supply network are selected according to the node selection principles, and flow and pressure sensors are deployed; the sensors upload and interact with data based on HTTP / HTTPS and TCP protocols.

[0087] In this embodiment, in step S3, the historical basic data and the monitoring data are uploaded to the server through the data stream processing framework; the historical basic data and the monitoring data are preprocessed by the server to obtain preprocessed data;

[0088] Specifically, a data stream processing framework such as Kafka is used to collect the historical basic data and the monitoring data and upload them to a deployed server, and the data is preprocessed as follows:

[0089] For the historical basic data, the hourly data and the temperature and humidity data, holiday information, and weekday data of the corresponding time node are stored as a data sequence;

[0090] For the monitoring data, the data of the hour and the predicted data obtained by the rural pipe network water pressure prediction model are stored as a data sequence. If there is no predicted data, the default value None is taken;

[0091] For abnormal data in the monitoring data, if the abnormal situation is the missing of some parameter values, the average of the corresponding data of the six hours before and after the hour is taken for filling; if the abnormal situation is the long-term repetition of parameter values, the data of the same time period of the same month is taken, multiplied by the deviation coefficient and used for filling, and the deviation coefficient is taken as the divisor average of the non-repeating data of the same time period of the same month and the non-repeating data of the repeated time period; if the abnormal situation is the parameter value fluctuation error, the average value of the data of the same time period in recent days is taken as filling.

[0092] In this embodiment, in step S4, a relational database management system is constructed based on MySQL; through the relational database management system, Hadoop and Spark tools are used to batch process the preprocessed data to obtain batch-processed historical basic data and monitoring data.

[0093] In this embodiment, in step S5, a rural pipe network water pressure prediction model is constructed; the rural pipe network water pressure prediction model is trained using the historical basic data to obtain the trained rural pipe network water pressure prediction model; the batch-processed historical basic data and monitoring data are input into the trained rural pipe network water pressure prediction model to obtain rural pipe network water pressure prediction data through prediction;

[0094] Specifically, for each calculation unit, the expression of the rural pipe network water pressure prediction model is:

[0095] pre.=αTime+βSpace+γH+σQ+ε

[0096] Where pre. is the predicted value of a node; Time is the temporal inertia term of the node, including the average water pressure of the node in the previous time period, the average water pressure of the same month and week, the average water pressure of the same year and quarter, and the average water pressure of the same quarter and week in the previous three years; Space is the spatial inertia term of the node, including the monitored water pressure of the upstream node in the same time period and the predicted water pressure of the upstream node in the same time period (this term does not exist for the initial calculation node); H is the benchmark term of the node, that is, the actual monitored water pressure of the node at the same time; Q is the water pressure correction term of the node, including the water volume monitoring values ​​of all nodes at the same time; ε is the bias term of the node, which is used to explain possible unconsidered rules in model learning; α, β, γ, and δ are the learning rates of the LSTM model, which are calibrated and corrected by the hyperparameter total lr and model learning.

[0097] Based on historical data, we use the data for each hour, the data for the previous six hours, and the data for the same week of the current month as input, and the data for the next hour as the output label. We then create a dataset, randomly selecting 2 / 3 of the dataset as the training set and the remaining 1 / 3 as the validation set. Each independent dataset contains at least one month's worth of data.

[0098] The rural pipe network water pressure prediction model is trained using the training set; the robustness and accuracy of the rural pipe network water pressure prediction model are verified using the validation set. After the validation is completed, the trained rural pipe network water pressure prediction model is obtained. The batch-processed historical basic data and monitoring data are input into the trained rural pipe network water pressure prediction model to obtain rural pipe network water pressure prediction data through prediction.

[0099] In this embodiment, in step S6, a pipe burst detection and water pressure control model is constructed; the batch-processed monitoring data and the rural pipe network water pressure prediction data are input into the pipe burst detection and water pressure control model; the pipe burst detection and water pressure control model detects and processes the data, outputs a list of high-risk pipe sections, and performs PID control;

[0100] like Figure 3 As shown, the detection and processing steps of the pipe burst detection and water pressure control model are as follows:

[0101] S61. Identify the batch-processed monitoring data and the rural pipe network water pressure prediction data using a decision tree model, output the pipe burst risk, and mark the pipe burst risk;

[0102] Among them, the Pytorch library based on Python language implements the definition, training and verification of decision tree models;

[0103] Specifically, data is extracted from batch-processed monitoring data and the rural pipe network water pressure prediction data. A decision tree model is trained based on historical abnormal pipe burst data to identify pipe burst risks. Input features include water pressure, flow rate, and historical abnormal patterns. The output value is a floating-point number between 0 and 1, representing the pipe burst risk. A confusion matrix and classification report are used to evaluate model performance, focusing on precision, recall, and F1 score. If a pipe burst risk is detected, the data is marked as "risk" and uploaded to monitoring personnel via a log. If there is no risk, the data is marked as "safe."

[0104] S62. Optimizing low-explosion pipe risk data through fuzzy control strategy;

[0105] Specifically,

[0106] (1) Determine input and output:

[0107] Input variables: water pressure error (the difference between the current water pressure and the target water pressure), rate of change of water pressure error (the speed at which the error changes);

[0108] Output variable: head adjustment amount.

[0109] (2) Define fuzzy sets:

[0110] Water pressure error:

[0111] Negative big (NB), negative small (NS), zero (ZE), positive small (PS), positive big (PB);

[0112] Water pressure error change rate:

[0113] Negative big (NB), negative small (NS), zero (ZE), positive small (PS), positive big (PB);

[0114] Head adjustment:

[0115] Decrease large (RD), decrease small (RS), remain unchanged (ZE), increase small (IS), increase large (ID).

[0116] (3) Fuzzy control rules:

[0117] If the water pressure error is positive large (PB) and the rate of change is positive small (PS), the head adjustment amount is increase large (ID);

[0118] If the water pressure error is small negative (NS) and the rate of change is large negative (NB), the head adjustment is small reduction (RS);

[0119] If the water pressure error is positive (PB) and the rate of change is zero (ZE), the head adjustment amount is incremental (ID);

[0120] If the water pressure error is positive small (PS) and the rate of change is positive small (PS), the head adjustment amount is increase small (IS);

[0121] If the water pressure error is negative (NB) and the rate of change is negative (NS), the head adjustment is reduced (RD);

[0122] If the water pressure error is zero (ZE) and the rate of change is zero (ZE), the head adjustment amount is constant (ZE);

[0123] If the water pressure error is negative small (NS) and the rate of change is zero (ZE), the head adjustment amount is reduced small (RS);

[0124] If the water pressure error is positive (PS) and the rate of change is zero (ZE), the head adjustment amount is increasing (IS).

[0125] (4) Membership function:

[0126] The Sigmoid function is used to define the membership:

[0127]

[0128] Adjust parameters a and b according to your needs.

[0129] (5) Fuzzy reasoning:

[0130] Use the minimum method (Min) to calculate the activation degree of the rules: for each rule, calculate the membership of the input conditions and take the minimum value as the activation degree of the rule.

[0131] (6) Defuzzification:

[0132] Use the mean method to defuzzify:

[0133] Z=∑μi*∑(μi·zi)

[0134] Where μi is the activation degree of each rule and zi is the output value of the rule.

[0135] S63: Time monitoring and processing of high-explosive pipe risk data; if the set time is exceeded, the log is automatically uploaded and the monitoring personnel are notified through the set method;

[0136] Specifically, for high-explosive pipe risk data, if the duration exceeds 5 minutes, the log will be automatically uploaded and the monitoring personnel will be notified via SMS or other means.

[0137] S64. Analyze abnormal water pressure nodes and the possibility of pipe burst, output a list of high-risk pipe sections and perform PID control.

[0138] Specifically,

[0139] (1) Data preparation:

[0140] Get the risk level of each node (0 to 1, with the risk level increasing as the number increases). Get the topological structure between nodes (directed edge weight graph).

[0141] (2) Abnormal node identification:

[0142] A risk threshold (0.4) is set, and risk nodes exceeding this threshold are identified and marked as abnormal nodes. In particular, data with a risk above 0.7 is marked as high-risk data.

[0143] (3) Pipeline section assessment:

[0144] For each abnormal node, obtain all its connected upstream and downstream nodes. Calculate the risk probability of adjacent pipe sections: If the risk level of both upstream and downstream nodes is high, the pipe section has a higher risk of burst. Consider the pipe section's historical failure history (if any) and increase its weight.

[0145] (4) Assessment of the possibility of pipe burst:

[0146] Calculate a burst probability score for each pipe segment. The score can be based on the average or maximum risk level of the nodes. Incorporate historical failure weights (if any). Identify the pipe segments with the highest scores as high-risk segments.

[0147] (5) Result output:

[0148] Upload a list of identified high-risk pipe sections to the logging system. The output format includes information such as pipe section identification, connection nodes, and burst probability score.

[0149] (6) Decision support:

[0150] Provide maintenance personnel with information on high-risk pipeline sections, prioritizing inspection and maintenance.

[0151] In this embodiment, in step S7, physical data of the rural water supply network is collected; based on the physical data, a geometric model of the rural water supply network is constructed based on ArcGIS;

[0152] Specifically, information on rural pipe network layout, pipe materials, pipe diameters, pipe lengths, and nodes is collected. Based on the collected physical information of the rural pipe network, a geometric model of the pipe network is established using ArcGIS:

[0153] Store relevant information in CSV file format, including node ID, X coordinate, and Y coordinate. Use the "Add Data" function to import the CSV file into the project, right-click the imported table, and select "Display XY Data". Specify the coordinate columns corresponding to the X and Y fields, and set the coordinate system (usually WGS 84 or a projection suitable for local conditions). Use the "Point to Line" tool (requires an advanced license for ArcGIS Pro or ArcMap). Specify the point layer as input and generate lines according to the node connection sequence. If there is no connection sequence, edit it manually or create it based on the topology information. Use the "Edit" tool to manually adjust the pipeline connection to ensure that it conforms to the actual pipeline network topology. You can add attribute fields such as pipe diameter, pipe material, etc. Use the "Topology" tool to check and fix topology errors, such as hanging nodes, overlapping line segments, etc. Set line symbols as needed. Add labels to display node or segment information.

[0154] In this embodiment, in step S8, a rural water supply network system data model is constructed based on the rural water supply network water pressure prediction model, the pipe burst detection and water pressure control model, and the rural water supply network geometric model;

[0155] Specifically, based on the rural pipe network water pressure prediction model, the pipe burst detection and water pressure control model and the rural water supply pipe network geometric model, the collected historical data and real-time monitored operation data are integrated into the rural water supply pipe network geometric model to construct a rural water supply pipe network system data model.

[0156] In this embodiment, in step S9, the back-end program of the rural water supply network system data model is written in Python language, and the rural water supply network system data model outputs time series curves of real-time data and predicted data to realize the visualization of the operating status of the water supply network.

[0157] Specifically, the back-end program is written in Python to generate time series curves of real-time data and predicted data, display the changing trend of data over time, provide a user interactive interface to facilitate viewing and analysis of data, and realize the visualization of the operating status of the water supply network.

[0158] Generate time series curves of real-time data, which can be Figure 4 As shown, showing abnormal conditions; able to Figure 5 As shown in the figure, the actual water pressure changes with time; Figure 6 As shown in , the predicted water pressure changes over time; Figure 7 As shown in the figure, the comparison between real-time data and forecast data is displayed through a visual interface to help users analyze and make decisions. Figure 8As shown in , it shows the changes of control parameters over time and provides a visual interface for users to supervise the fuzzy control process; Figure 9 As shown, the process of eliminating abnormal water pressure in the pipeline network is demonstrated.

[0159] In summary, the present invention collects the set data of the rural prediction area to obtain the historical basic data of the rural prediction area; selects the nodes of the rural water supply network according to the node selection principle; deploys flow and pressure sensors at the selected nodes, and obtains monitoring data through the flow and pressure sensors; uploads the historical basic data and the monitoring data to the server through the data stream processing framework; pre-processes the historical basic data and the monitoring data through the server to obtain pre-processed data; builds a relational database management system based on MySQL; uses Hadoop and Spark tools to batch-process the pre-processed data through the relational database management system to obtain batch-processed historical basic data and monitoring data; builds a rural pipe network water pressure prediction model; trains the rural pipe network water pressure prediction model through the historical basic data to obtain the trained rural pipe network water pressure prediction model; and uploads the batch-processed historical basic data and the monitoring data to the server. The basic data and monitoring data are input into the trained rural pipe network water pressure prediction model, and the rural pipe network water pressure prediction data is obtained through prediction; a pipe burst detection and water pressure control model is constructed; the batch-processed monitoring data and the rural pipe network water pressure prediction data are input into the pipe burst detection and water pressure control model, and the pipe burst detection and water pressure control model is used for detection and processing, and a list of high-risk pipe sections is output and PID control is performed; the physical data of the rural water supply network is collected; according to the physical data, a rural water supply network geometric model is constructed based on ArcGIS; according to the rural pipe network water pressure prediction model, the pipe burst detection and water pressure control model and the rural water supply network geometric model, a rural water supply network system data model is constructed; the back-end program of the rural water supply network system data model is written in Python language, and the rural water supply network system data model outputs time series curves of real-time data and prediction data to realize visualization of the operation status of the water supply network. This invention analyzes historical data, mines data features, and constructs a water pressure prediction model based on digital twins. This allows for precise monitoring and early warning of water pressure across the entire water supply network, ensuring water supply safety. This invention fully considers the characteristics of rural water supply and water usage, enabling intelligent management and control of rural water supply systems, effectively ensuring the reliability and safety of rural water supply.

[0160] It should be noted that the method of the embodiments of the present disclosure can be performed by a single device, such as a computer or server. The method of the embodiments of the present disclosure can also be applied in a distributed scenario, where multiple devices cooperate to perform the method. In such a distributed scenario, one of the multiple devices may only perform one or more steps of the method of the embodiments of the present disclosure, and the multiple devices will interact with each other to complete the method.

[0161] It should be noted that the above description is limited to some embodiments of the present disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in an order different from that described in the above embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0162] Example 2

[0163] See also Figure 10 , Embodiment 2 of the present invention also provides a rural water supply network pressure monitoring and analysis device based on digital twins, including:

[0164] The historical basic data collection module 001 is used to collect the set data of the rural prediction area and obtain the historical basic data of the rural prediction area;

[0165] Monitoring data acquisition module 002 is used to select nodes of the rural water supply network according to the node selection principle; deploy flow and pressure sensors at the selected nodes and obtain monitoring data through the flow and pressure sensors;

[0166] The data preprocessing module 003 is used to upload the historical basic data and the monitoring data to the server through the data stream processing framework; and preprocess the historical basic data and the monitoring data through the server to obtain preprocessed data;

[0167] The data batch processing module 004 is used to build a relational database management system based on MySQL; through the relational database management system, the pre-processed data is batch-processed using Hadoop and Spark tools to obtain batch-processed historical basic data and monitoring data;

[0168] The rural pipe network water pressure prediction model construction and processing module 005 is used to construct a rural pipe network water pressure prediction model; train the rural pipe network water pressure prediction model using the historical basic data to obtain the trained rural pipe network water pressure prediction model; input the batch-processed historical basic data and monitoring data into the trained rural pipe network water pressure prediction model to obtain rural pipe network water pressure prediction data through prediction;

[0169] The pipe burst detection and water pressure control model construction and processing module 006 is used to construct a pipe burst detection and water pressure control model; the batch-processed monitoring data and the rural pipe network water pressure prediction data are input into the pipe burst detection and water pressure control model; the pipe burst detection and water pressure control model detects and processes the data, outputs a list of high-risk pipe sections, and performs PID control;

[0170] The rural water supply network geometric model construction module 007 is used to collect physical data of the rural water supply network; and construct a rural water supply network geometric model based on ArcGIS according to the physical data;

[0171] A rural water supply network system data model construction module 008 is used to construct a rural water supply network system data model based on the rural water supply network water pressure prediction model, the pipe burst detection and water pressure control model, and the rural water supply network geometric model;

[0172] The rural water supply network system data model visualization output module 009 is used to write the back-end program of the rural water supply network system data model using the Python language. The rural water supply network system data model outputs time series curves of real-time data and predicted data to realize the visualization of the operating status of the water supply network.

[0173] In this embodiment, in the historical basic data collection module 001, in the process of collecting the set data for the rural prediction area, the set data includes: historical water supply and actual water use hourly data, temperature and humidity data at corresponding time nodes, data on the day of the week corresponding to the time of the day, and numerical data on whether the day is a holiday.

[0174] In this embodiment, in the monitoring data acquisition module 002, in the process of selecting nodes of the rural water supply network according to the node selection principle, the node selection principle includes: laying at user entry points, network intersections, and at intervals on straight sections of the network.

[0175] In this embodiment, in the data preprocessing module 003, during the preprocessing of the historical basic data and the monitoring data by the server, the historical basic data is stored as a data sequence along with the temperature and humidity data, holiday information, and weekday data of the corresponding time node.

[0176] For the monitoring data, the data of the hour and the predicted data obtained by the rural pipe network water pressure prediction model are stored as a data sequence. If there is no predicted data, the default value None is taken;

[0177] For abnormal data in the monitoring data, if the abnormal situation is the missing of some parameter values, the average of the corresponding data of the six hours before and after the hour is taken for filling; if the abnormal situation is the long-term repetition of parameter values, the data of the same time period of the same month is taken, multiplied by the deviation coefficient and used for filling, and the deviation coefficient is taken as the divisor average of the non-repeating data of the same time period of the same month and the non-repeating data of the repeated time period; if the abnormal situation is the parameter value fluctuation error, the average value of the data of the same time period in recent days is taken as filling.

[0178] In this embodiment, in the rural pipe network water pressure prediction model construction and processing module 005, the expression of the rural pipe network water pressure prediction model is:

[0179] pre.=αTime+βSpace+γH+σQ+ε

[0180] Where pre. is the predicted value of a node; Time is the temporal inertia term of the node, including the average water pressure of the node in the previous time period, the average water pressure of the same month and week, the average water pressure of the same year and quarter, and the average water pressure of the same quarter and week in the previous three years; Space is the spatial inertia term of the node, including the monitored water pressure of the upstream node in the same time period and the predicted water pressure of the upstream node in the same time period (this term does not exist for the initial calculation node); H is the benchmark term of the node, that is, the actual monitored water pressure of the node at the same time; Q is the water pressure correction term of the node, including the water volume monitoring values ​​of all nodes at the same time; ε is the bias term of the node, which is used to explain possible unconsidered rules in model learning; α, β, γ, and δ are the learning rates of the LSTM model, which are calibrated and corrected by the hyperparameter total lr and model learning.

[0181] In this embodiment, in the pipe burst detection and water pressure control model construction and processing module 006, the detection and evaluation submodules include:

[0182] The data identification and marking submodule 061 is used to identify the batch-processed monitoring data and the rural pipe network water pressure prediction data through a decision tree model, output the pipe burst risk, and mark the pipe burst risk;

[0183] The low-explosion pipe risk data optimization submodule 062 is used to optimize the low-explosion pipe risk data through a fuzzy control strategy;

[0184] High-explosive pipe risk data processing submodule 063 is used to monitor and process high-explosive pipe risk data over time; if the set time is exceeded, the log is automatically uploaded and the monitoring personnel are notified through the set method;

[0185] The possibility analysis submodule 064 is used to analyze abnormal water pressure nodes and the possibility of pipe burst, output a list of high-risk pipe sections and perform PID control.

[0186] It should be noted that the information interaction, execution process, etc. between the modules of the above-mentioned system are based on the same concept as the method embodiment in Example 1 of the present application, and the technical effects they bring are the same as those of the method embodiment of the present application. For specific contents, please refer to the description in the method embodiment shown above in the present application, and no further details will be given here.

[0187] Example 3

[0188] Example 3 of the present invention provides a non-transitory computer-readable storage medium, which stores a program code for a rural water supply network pressure monitoring and analysis method based on digital twins. The program code includes instructions for executing Example 1 or any possible implementation method of the rural water supply network pressure monitoring and analysis method based on digital twins.

[0189] Computer-readable storage media can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state drives (SSDs)).

[0190] Example 4

[0191] Embodiment 4 of the present invention provides an electronic device, including: a memory and a processor;

[0192] The processor and the memory communicate with each other through a bus; the memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute the rural water supply network pressure monitoring and analysis method based on digital twins of Example 1 or any possible implementation thereof.

[0193] Specifically, the processor can be implemented by hardware or by software. When implemented by hardware, the processor can be a logic circuit, an integrated circuit, etc.; when implemented by software, the processor can be a general-purpose processor, which is implemented by reading software code stored in a memory. The memory can be integrated into the processor or located outside the processor and exist independently.

[0194] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable systems. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode.

[0195] Obviously, those skilled in the art will appreciate that the various modules or steps of the present invention described above can be implemented using a general-purpose computing system. They can be centralized on a single computing system or distributed across a network of multiple computing systems. Alternatively, they can be implemented using program code executable by a computing system, and thus, they can be stored in a storage system and executed by the computing system. In some cases, the steps shown or described herein can be performed in a different order than that shown, or they can be fabricated into separate integrated circuit modules, or multiple modules or steps can be fabricated into a single integrated circuit module. Thus, the present invention is not limited to any particular combination of hardware and software.

[0196] Although the present invention has been described in detail above using general descriptions and specific embodiments, it will be apparent to those skilled in the art that modifications and improvements may be made thereto. Therefore, such modifications and improvements, without departing from the spirit of the present invention, are intended to be within the scope of protection claimed herein.

Claims

1. A rural water supply network pressure monitoring and analysis method based on digital twins, characterized in that: include: Collect the setting data of the rural forecast area and obtain the historical basic data of the rural forecast area; According to the node selection principle, select the nodes of the rural water supply network; Deploy flow and pressure sensors at selected nodes, and obtain monitoring data through the flow and pressure sensors; Uploading the historical basic data and the monitoring data to the server through a data stream processing framework; Preprocessing the historical basic data and the monitoring data by the server to obtain preprocessed data; Building a relational database management system based on MySQL; using the relational database management system, using Hadoop and Spark tools to batch process the pre-processed data to obtain batch-processed historical basic data and monitoring data; Constructing a rural pipe network water pressure prediction model; training the rural pipe network water pressure prediction model through the historical basic data to obtain the trained rural pipe network water pressure prediction model; inputting the batch-processed historical basic data and monitoring data into the trained rural pipe network water pressure prediction model to obtain rural pipe network water pressure prediction data through prediction; Constructing a pipe burst detection and water pressure control model; inputting the batch-processed monitoring data and the rural pipe network water pressure prediction data into the pipe burst detection and water pressure control model, and outputting a list of high-risk pipe sections through detection and processing by the pipe burst detection and water pressure control model and performing PID control; Collect physical data of rural water supply networks; According to the physical data, a rural water supply network geometric model is constructed based on ArcGIS; Constructing a rural water supply network system data model based on the rural pipe network water pressure prediction model, the pipe burst detection and water pressure control model and the rural water supply network geometric model; The back-end program of the rural water supply network system data model is written in Python language. The rural water supply network system data model outputs time series curves of real-time data and predicted data to realize the visualization of the operation status of the water supply network.

2. The rural water supply network pressure monitoring and analysis method based on digital twin according to claim 1 is characterized in that: In the process of collecting the set data for the rural prediction area, the set data includes: historical water supply and actual water use hour data, temperature and humidity data at corresponding time nodes, data on the day of the week corresponding to the time of the day, and numerical data on whether the day is a holiday.

3. The rural water supply network pressure monitoring and analysis method based on digital twin according to claim 2 is characterized in that: In the process of selecting nodes of the rural water supply network according to the node selection principle, the node selection principle includes: laying at user entry points, network intersections, and at intervals on straight sections of the network.

4. The rural water supply network pressure monitoring and analysis method based on digital twin according to claim 3 is characterized in that: In the process of preprocessing the historical basic data and the monitoring data by the server, for the historical basic data, the data of the hour and the temperature and humidity data of the corresponding time node, the holiday information and the day of the week data are stored as a data sequence; For the monitoring data, the data of the hour and the predicted data obtained by the rural pipe network water pressure prediction model are stored as a data sequence. If there is no predicted data, the default value None is taken; For abnormal data in the monitoring data, if the abnormal situation is the missing of some parameter values, the average of the corresponding data of the six hours before and after the hour is taken for filling; if the abnormal situation is the long-term repetition of parameter values, the data of the same time period of the same month is taken, multiplied by the deviation coefficient for filling, and the deviation coefficient is the divisor average of the non-repeating data of the same time period of the same month and the non-repeating data of the repeated time period; if the abnormality is a fluctuation error of the parameter value, the average value of the data of the same time period in recent days is taken for filling.

5. The rural water supply network pressure monitoring and analysis method based on digital twin according to claim 4 is characterized in that: The expression of the rural pipe network water pressure prediction model is: pre.=αTime+βSpace+γH+σQ+ε Where pre. is the predicted value of a certain node; Time is the time inertia term of the node, including the average water pressure in the previous time period, the average water pressure in the same month and week, the average water pressure in the same quarter of the same year, and the average water pressure in the same quarter and week of the previous three years; Space is the spatial inertia term of the node, including the monitored water pressure of the upstream node in the same time period and the predicted water pressure of the upstream node in the same time period (this item does not exist for the initial calculation node); H is the benchmark term of the node, that is, the actual monitored water pressure of the node at the same time; Q is the water pressure correction term of the node, including the water volume monitoring values ​​of all nodes at the same time; ε is the bias term of the node, which is used to explain possible unconsidered regularities in model learning; α, β, γ, and δ are the learning rates of each item in the LSTM model, which are corrected by the hyperparameter total lr and model learning.

6. The rural water supply network pressure monitoring and analysis method based on digital twin according to claim 5 is characterized in that: The detection and processing steps of the pipe burst detection and water pressure control model are as follows: The batch-processed monitoring data and the rural pipe network water pressure prediction data are identified by a decision tree model, the pipe burst risk is output, and the pipe burst risk is marked; Optimize low pipe burst risk data through fuzzy control strategy; Conduct time monitoring and processing of high-explosive pipe risk data; If the set time is exceeded, the log will be automatically uploaded and the monitoring personnel will be notified through the set method; Analyze abnormal water pressure nodes and the possibility of pipe burst, output a list of high-risk pipe sections and perform PID control.

7. A rural water supply network pressure monitoring and analysis device based on digital twin, adopting a rural water supply network pressure monitoring and analysis method based on digital twin according to any one of claims 1 to 6, characterized in that: include: A historical basic data collection module is used to collect the set data of the rural prediction area and obtain the historical basic data of the rural prediction area; A monitoring data acquisition module is used to select nodes of the rural water supply network according to the node selection principle; deploy flow and pressure sensors at the selected nodes, and acquire monitoring data through the flow and pressure sensors; A data preprocessing module, used for uploading the historical basic data and the monitoring data to a server through a data stream processing framework; Preprocessing the historical basic data and the monitoring data by the server to obtain preprocessed data; A data batch processing module is used to build a relational database management system based on MySQL; through the relational database management system, the pre-processed data is batch processed using Hadoop and Spark tools to obtain batch-processed historical basic data and monitoring data; A rural pipe network water pressure prediction model construction and processing module is used to construct a rural pipe network water pressure prediction model; train the rural pipe network water pressure prediction model through the historical basic data to obtain the trained rural pipe network water pressure prediction model; input the batch-processed historical basic data and monitoring data into the trained rural pipe network water pressure prediction model to obtain rural pipe network water pressure prediction data through prediction; A pipe burst detection and water pressure control model construction and processing module is used to construct a pipe burst detection and water pressure control model; the batch-processed monitoring data and the rural pipe network water pressure prediction data are input into the pipe burst detection and water pressure control model, and the pipe burst detection and water pressure control model is used for detection and processing, and a high-risk pipe section list is output and PID control is performed; A rural water supply network geometric model building module is used to collect physical data of the rural water supply network; According to the physical data, a rural water supply network geometric model is constructed based on ArcGIS; A rural water supply network system data model construction module, used to construct a rural water supply network system data model based on the rural network water pressure prediction model, the pipe burst detection and water pressure control model and the rural water supply network geometric model; The rural water supply network system data model visualization output module is used to write the back-end program of the rural water supply network system data model in Python language. The rural water supply network system data model outputs time series curves of real-time data and predicted data to realize the visualization of the operation status of the water supply network.

8. The rural water supply network pressure monitoring and analysis device based on digital twin according to claim 7 is characterized in that: In the data preprocessing module, in the process of preprocessing the historical basic data and the monitoring data by the server, for the historical basic data, the data of the hour and the temperature and humidity data of the corresponding time node, the holiday information and the day of the week data are stored as a data sequence; For the monitoring data, the data of the hour and the predicted data obtained by the rural pipe network water pressure prediction model are stored as a data sequence. If there is no predicted data, the default value None is taken; For abnormal data in the monitoring data, if the abnormal situation is the missing of some parameter values, the average of the corresponding data of the six hours before and after the hour is taken for filling; if the abnormal situation is the long-term repetition of parameter values, the data of the same time period of the same month is taken, multiplied by the deviation coefficient for filling, and the deviation coefficient is the divisor average of the non-repeating data of the same time period of the same month and the non-repeating data of the repeated time period; if the abnormality is a fluctuation error of the parameter value, the average value of the data of the same time period in recent days is taken for filling.

9. The rural water supply network pressure monitoring and analysis device based on digital twin according to claim 8 is characterized in that: In the rural pipe network water pressure prediction model construction and processing module, the expression of the rural pipe network water pressure prediction model is: pre.=αTime+βSpace+γH+σQ+ε Where pre. is the predicted value of a certain node; Time is the time inertia term of the node, including the average water pressure in the previous time period, the average water pressure in the same month and week, the average water pressure in the same quarter of the same year, and the average water pressure in the same quarter and week of the previous three years; Space is the spatial inertia term of the node, including the monitored water pressure of the upstream node in the same time period and the predicted water pressure of the upstream node in the same time period (this item does not exist for the initial calculation node); H is the benchmark term of the node, that is, the actual monitored water pressure of the node at the same time; Q is the water pressure correction term of the node, including the water volume monitoring values ​​of all nodes at the same time; ε is the bias term of the node, which is used to explain possible unconsidered regularities in model learning; α, β, γ, and δ are the learning rates of each item in the LSTM model, which are corrected by the hyperparameter total lr and model learning.

10. The rural water supply network pressure monitoring and analysis device based on digital twin according to claim 9 is characterized in that: In the pipe burst detection and water pressure control model construction and processing module, the detection and evaluation submodules include: A data identification and marking submodule is used to identify the batch-processed monitoring data and the rural pipe network water pressure prediction data through a decision tree model, output the pipe burst risk, and mark the pipe burst risk; The low-explosion pipe risk data optimization submodule is used to optimize the low-explosion pipe risk data through fuzzy control strategy; The high-explosive pipe risk data processing submodule is used to monitor and process the high-explosive pipe risk data over time; if the set time is exceeded, the log is automatically uploaded and the monitoring personnel are notified in a set manner; The possibility analysis submodule is used to analyze abnormal water pressure nodes and the possibility of pipe burst, output a list of high-risk pipe sections and perform PID control.

Images