Urban pipeline leakage monitoring system based on DMA partition management technology

Abstract

The invention relates to the field of water supply systems, and discloses an urban pipeline leakage monitoring system based on a DMA partition management technology, and the system comprises a data collection module which is used for collecting the data of a flowmeter, a pressure sensor and a hydrophone in a DMA partition; the data preprocessing module is used for preprocessing the data; the time sequence analysis module is used for performing seasonal and trend decomposition and nonlinear time sequence analysis on the preprocessed data; the leakage detection module is used for predicting and detecting leakage; the noise analyzing and positioning module is used for analyzing and determining the specific position of the leakage point of the pipe section through the noise data of the hydrophone; and the real-time prediction alarm module is used for processing data in real time and triggering alarm when leakage is detected. The system not only can predict a potential leakage event in advance, but also can quickly position and give an alarm when leakage occurs, so that the operation efficiency and safety of the water supply system are greatly improved.

Description

Technical Field

[0001] The invention relates to the technical field of water supply systems, and in particular to an urban pipeline leakage monitoring system based on DMA zoning management technology. Background Art

[0002] Urban water supply network is an important part of urban infrastructure. During the long-term operation of water supply network, leakage often occurs due to various factors such as natural aging, corrosion, external damage, construction quality, etc. Leakage not only wastes precious water resources, but also may cause serious problems such as ground collapse and water supply interruption, which have a negative impact on the normal operation of the city and the lives of residents.

[0003] Traditional methods for detecting water supply network leakage mainly rely on manual inspection and passive detection. Manual inspection methods are time-consuming and labor-intensive, inefficient, and difficult to detect and repair leakage in a timely manner. Passive detection methods, such as acoustic wave detection and infrared detection, have improved the efficiency of leakage detection to a certain extent, but still have problems such as limited detection range, low accuracy, and high environmental interference. In addition, these methods can often only detect leakage after it has occurred and caused certain consequences, and cannot achieve early warning and active prevention and control.

[0004] In recent years, with the rapid development of the Internet of Things (IoT), big data and artificial intelligence technologies, intelligent water supply network monitoring systems have gradually become a hot topic in research and application. These systems collect data such as flow, pressure, noise, etc. in real time by arranging various sensors in the pipe network, and use advanced data analysis and processing technologies to monitor and analyze the operation status of the pipe network in real time, thereby achieving early detection and positioning of leakage. However, the existing intelligent monitoring systems still have some problems and challenges in practical applications. First, the complexity and computational complexity of data processing and analysis algorithms are large, which may lead to insufficient real-time performance. Secondly, the accuracy of leakage detection and positioning needs to be further improved, especially in complex urban water supply pipe networks, how to accurately and quickly locate the leakage point is still a difficult problem. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention provides an urban pipeline leakage monitoring system based on DMA zoning management technology, which realizes the early detection and precise positioning of urban water supply network leakage through efficient and accurate monitoring and analysis.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: an urban pipeline leakage monitoring system based on DMA zoning management technology, comprising:

[0007] Data collection module, used to collect data from flow meters, pressure sensors and hydrophones in DMA zones;

[0008] A data preprocessing module, used for preprocessing the data;

[0009] Time series analysis module, used to perform seasonality and trend decomposition and nonlinear time series analysis on preprocessed data;

[0010] The leakage detection module includes a random forest regression model and a density-based local anomaly factor detection algorithm for predicting and detecting leakage;

[0011] The noise analysis and positioning module is used to determine the specific location of the pipe leak through the noise data analysis of the hydrophone;

[0012] Real-time prediction and alarm module, which processes data in real time and triggers alarms when leakage is detected.

[0013] Preferably, the data collection module is used to collect data from various sensor devices, and the sensors include:

[0014] Flow sensors are located at the inlet and outlet of each DMA partition to measure the water flow through a specific pipe section to monitor the water flow dynamics in the area. The data types include instantaneous flow and accumulated flow.

[0015] Pressure sensors are installed at key pipe sections, including zone inlets, outlets and potential high-risk areas, to monitor water pressure changes in real time and facilitate early identification of abnormal pressure fluctuations in the system;

[0016] Hydrophones are deployed in pipe sections near known and potential leaks to capture sound signals caused by water flow for subsequent leak location analysis;

[0017] Auxiliary sensing equipment, including but not limited to temperature sensors and water quality monitoring equipment, is used to provide environmental and water quality data to support the analysis of environmental factors for leakage detection.

[0018] Preferably, the data preprocessing module includes:

[0019] Data cleaning unit to remove data during non-working hours, correct obvious erroneous readings, and fill in missing values;

[0020] Data normalization unit, used to normalize data from different sources and scopes to a uniform scale;

[0021] The feature engineering unit is used to extract key features from the raw data that are helpful for leakage detection and prediction, including extracting statistical features, time features, and frequency domain features from flow and pressure time series data.

[0022] Preferably, the time series analysis module includes:

[0023] The seasonality and trend decomposition unit is used to perform seasonality and trend decomposition on time series data to identify cyclical and long-term change patterns in the data. The decomposition is performed using the following formula:

[0024] y t =T t +S t +R t

[0025] Among them, y t represents the observed value at time point t, T t is the trend component, S t is the seasonal component, R t is the remaining amount;

[0026] The nonlinear time series analysis unit uses a nonlinear autoregressive moving average model to simulate and predict time series data using the following model formula:

[0027]

[0028] Among them, c is a constant term, φ i is the autoregressive coefficient, θ j is the moving average coefficient, ∈ t is the error term at time point t, and p and q represent the orders of the autoregressive and moving average models, respectively.

[0029] Preferably, the leakage detection module includes a random forest regression model for predicting leakage probability and impact range, and the model is trained with historical data and calculates the importance of each feature.

[0030] Preferably, the local abnormal factor detection algorithm of the leakage detection module includes:

[0031] The local reachability density is calculated for each observation point to measure the local density of the observation point. The formula is as follows:

[0032]

[0033] Where x is the current data point, N k (x) is the set of k nearest neighbors of point x, reach-dist k (x,o) is the reachable distance between point x and its neighbor o;

[0034] Based on the above local reachable density, the local anomaly factor is calculated for each data point to identify the degree of anomaly relative to its neighboring points. The formula is as follows:

[0035]

[0036] Among them, lrdk (o) is the local reachability density of neighboring point o, lrd k (x) is the local reachable density of data point x. The larger the ratio is, the more likely point x is an outlier.

[0037] Preferably, the noise analysis and positioning module includes:

[0038] The noise feature extraction unit uses Fourier transform to convert the noise signal in the time domain to the frequency domain for analysis. The formula is as follows:

[0039]

[0040] Where x(t) represents the noise signal strength at time t, f is the frequency, and X(f) is the complex frequency domain representation of the corresponding frequency f;

[0041] The noise pattern recognition unit uses an improved K-means clustering algorithm to classify the noise data in the frequency domain to identify specific water leakage noise patterns. First, the distance from each noise sample to each class center is calculated, and then the class center is updated. The iteration is repeated until the convergence condition is met;

[0042] The leak location unit uses the noise data collected by multiple hydrophones installed at different locations to locate the leak through triangulation, using the following formula:

[0043]

[0044] Among them, (x i ,y i ) is the coordinate of the leakage point signal recorded by the ith hydrophone, n is the number of hydrophones, and Location represents the average coordinate of the leakage point calculated based on the collected noise data.

[0045] Preferably, the real-time prediction alarm module includes:

[0046] Real-time data processing unit, which uses stream processing framework to clean, process and analyze data;

[0047] The alarm system triggers an alarm when the predicted leakage probability or LOF score exceeds the threshold, and notifies maintenance personnel via SMS, email or APP.

[0048] The present invention also provides a method for monitoring urban pipeline leakage based on DMA zoning management technology, comprising the following steps:

[0049] Collect data from flow meters, pressure sensors and hydrophones within the DMA zone;

[0050] Cleaning, standardizing and feature engineering the data;

[0051] Perform seasonality and trend decomposition and nonlinear time series analysis on preprocessed data;

[0052] Predict and detect leakage through random forest regression model and density-based local anomaly factor detection algorithm;

[0053] The noise data from the hydrophone is used to extract noise features and perform pattern recognition through Fourier transform and improved K-means clustering algorithm.

[0054] Use triangulation to locate leaks;

[0055] Process data in real time and trigger an alarm when a leak is detected, notifying maintenance personnel via SMS, email or APP.

[0056] The present invention also provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the above method is implemented.

[0057] The present invention also provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the method described above is implemented.

[0058] The present invention provides a city pipeline leakage monitoring system based on DMA zoning management technology. It has the following beneficial effects:

[0059] 1. The present invention can efficiently and accurately predict and detect leakage events by introducing a random forest regression model and a density-based local anomaly factor detection algorithm. The random forest regression model is trained using historical data and can accurately predict future flow and pressure changes, providing a reliable basis for leakage prediction. The local anomaly factor detection algorithm identifies potential anomalies in real time by analyzing the local density differences of data points, effectively reducing the occurrence of false positives and false negatives, thereby significantly improving the accuracy of leakage detection.

[0060] 2. The real-time prediction alarm module of the present invention can immediately trigger an alarm and notify maintenance personnel when a leakage event is detected. Alarm information is sent through SMS, email, or APP to ensure that maintenance personnel can receive notifications and take actions in a timely manner. This real-time alarm mechanism can not only issue an early warning as soon as a leakage occurs, reducing the waste of water resources and the risk of pipeline damage, but also greatly improve the response speed and maintenance efficiency of the system.

[0061] 3. The present invention can utilize the noise data of the hydrophone through the noise analysis and positioning module, extract noise features and perform pattern recognition through Fourier transform and improved K-means clustering algorithm, and accurately calculate the leak location by combining triangulation. The combination of multi-point noise collection and triangulation enables the system to accurately locate the leak point and quickly guide maintenance personnel to repair it, reducing unnecessary excavation and maintenance costs and improving the efficiency and accuracy of pipeline maintenance.

[0062] 4. The time series analysis module of the present invention reveals the inherent laws of the data through seasonal and trend decomposition of the data and nonlinear time series analysis, and provides a reliable prediction basis for leakage detection. Seasonal and trend decomposition help identify cyclical changes and long-term trends in the data, while nonlinear time series analysis is used to model and predict complex time-dependent structures. This comprehensive data processing and analysis method enables the system to perform leakage detection and prediction more efficiently, improving the operational efficiency of the water supply system.

[0063] 5. The present invention can significantly reduce the waste of water resources and pipeline maintenance costs caused by leakage through efficient leakage detection and precise positioning. The real-time prediction and alarm mechanism ensures that leakage events are handled as soon as they occur, reducing the impact of leakage on the water supply system. In addition, accurate leakage point positioning and efficient maintenance guidance make maintenance work more targeted and efficient, further reducing maintenance costs and resource waste. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 It is a schematic diagram of DMA partition of the urban water supply network of the present invention;

[0065] Figure 2 It is a schematic diagram of the structure of the leakage monitoring system of the present invention;

[0066] Figure 3 It is a schematic diagram of the leakage monitoring method of the present invention;

[0067] Figure 4 It is a schematic diagram of the computer device structure of the present invention.

[0068] Among them, 40, computer equipment; 41, processor; 42, memory; 43, storage medium. DETAILED DESCRIPTION

[0069] The following will be combined with the drawings in the specification of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0070] The present invention relates to a leakage monitoring system for an urban water supply network based on DMA (area metering management) technology. DMA technology is an efficient method for reducing water loss and improving water efficiency management in urban water supply systems. The technology divides the urban water supply network into several relatively independent metering areas, namely DMA areas, each of which is equipped with inlet and outlet flow measurement equipment, so that each area can perform independent water balance calculations.

[0071] The core of DMA zoning is to effectively control and manage water loss in each area. In specific implementation, zoning planning is first carried out based on the structure of the water supply network, user distribution and historical leakage data. Through GIS (Geographic Information System) and other intelligent system tools, the logical and physical boundaries of DMA zoning are ensured to optimize water flow and pressure management. Figure 1 As shown, a schematic diagram of the DMA zoning of the urban water supply network is shown.

[0072] Based on DMA zoning, the pipeline leakage monitoring system of the present invention further develops a series of innovative monitoring and analysis technologies. The system combines modern sensing technology, data communication technology and data analysis technology to achieve real-time and accurate monitoring of the entire urban water supply network. The system collects key data for subsequent analysis and processing by installing flow meters, pressure sensors, hydrophones and noise recorders in each DMA area. The details are as follows:

[0073] Please see attached Figure 2 The embodiment of the present invention provides a city pipeline leakage monitoring system based on DMA zoning management technology, which comprehensively utilizes modern sensing, data processing and analysis technologies to monitor, predict and locate leakage in the city water supply network. The core components of the system include data collection module, data preprocessing module, time series analysis module, leakage detection module, noise analysis and positioning module and real-time prediction and alarm module.

[0074] Data collection module

[0075] Regarding the data collection module, first of all, it should be noted that data collection is the basis and key link of the entire urban pipeline leakage monitoring system. Only by accurately and timely collecting data from various sensor devices can the effectiveness of subsequent data preprocessing, analysis and prediction work be ensured. Therefore, the embodiment of the present invention designs an efficient data collection module, which aims to obtain detailed water flow, pressure and sound data from various sensors in the DMA partition, and provide reliable basic data support for leakage detection.

[0076] In this embodiment, the data collection module includes the following components:

[0077] Flow sensor: Flow meters are installed at the inlet and outlet of each DMA partition to measure water flow. Each flow meter can accurately record instantaneous flow and accumulated flow data. The data from the flow sensor is not only used to calculate the water balance within the partition, but also helps identify sudden leakage events.

[0078] Pressure sensors: Installed in key pipe sections, including partition inlets, outlets and potential high-risk areas, to monitor water pressure changes in real time. Data from pressure sensors helps identify abnormal pressure fluctuations in the system, which may be early signs of leakage. Pressure sensors can be used to detect and warn of possible leaks in a timely manner.

[0079] Hydrophone: Deployed in pipe sections near known and potential leaks to capture sound signals caused by water flow. Hydrophones can detect tiny leaking sounds in pipes. By analyzing these sound signals, the system can identify the specific location of the leak. The data collected by the hydrophone provides the basis for subsequent noise analysis and positioning modules.

[0080] Auxiliary sensing equipment: including temperature sensors, water quality monitoring equipment, etc., used to provide environmental and water quality data. These data are also very important for accurately identifying and locating leaks in some cases. For example, changes in water temperature may affect water pressure, and changes in water quality may be related to leaks.

[0081] These sensors transmit data to the central processing system via wired or wireless means. The data collection module must not only ensure the accuracy and timeliness of the data, but also have a certain fault tolerance to cope with possible sensor failures or delays in data transmission.

[0082] In this embodiment, the data of all sensors are aggregated through the IoT Gateway and transmitted to the central processing system in real time. In order to ensure the stability and reliability of data transmission, the system adopts a variety of transmission protocols (such as MQTT, HTTP, etc.) and protects the security of data through encryption. The central processing system performs preliminary storage and backup of the received data to prevent data loss.

[0083] In summary, the data collection module, as a basic component of the present invention, provides a solid foundation for subsequent leakage detection and analysis through accurate and efficient data collection and transmission. The coordinated work of various sensing devices enables the system to comprehensively and real-timely monitor the water flow, pressure and sound conditions within the DMA partition, ensuring the accuracy and timeliness of leakage detection.

[0084] Data preprocessing module

[0085] Regarding the data preprocessing module, it should be noted that data preprocessing is one of the key steps in the urban pipeline leakage monitoring system. The collected raw data usually contains noise, missing values ​​and outliers. Directly using these data for analysis and prediction may lead to inaccurate results. Therefore, the data preprocessing module ensures the quality and consistency of the data through a series of cleaning and standardization processes, laying a solid foundation for subsequent time series analysis and leakage detection.

[0086] In this embodiment, the data preprocessing module includes the following components:

[0087] Data cleaning unit: used to remove data during non-working hours, correct obvious erroneous readings and fill in missing values. The specific operations are as follows:

[0088] Filter data during non-working hours to ensure the timeliness and relevance of analytical data.

[0089] Fill missing values ​​using statistical methods such as mean filling, linear interpolation, or KNN (k nearest neighbors) algorithm.

[0090] Identify and remove outliers. Determine the abnormal threshold of the data through statistical analysis (such as the z-score method) and remove abnormal data points that exceed the threshold.

[0091] Data standardization unit: used to standardize data from different sources and ranges to a uniform scale to eliminate dimensional differences and make the data comparable. Specific methods include:

[0092] Min-Max Normalization: Scale the data to the interval [0,1].

[0093] Z-score standardization: transform the data into a standard normal distribution with a mean of 0 and a standard deviation of 1.

[0094] Feature Engineering Unit: It is used to extract key features from raw data that are helpful for leakage detection and prediction. The specific operations are as follows:

[0095] Extract statistical features: such as mean, standard deviation, maximum, minimum and median.

[0096] Extract time features: such as daily, weekly, and monthly periodic features to capture seasonal changes in the data.

[0097] Extract frequency domain features: Use Fourier transform (FFT) to convert time domain signals into frequency domain signals and analyze the spectral characteristics of the signals.

[0098] In this embodiment, the data preprocessing module not only ensures the accuracy and consistency of the data, but also enriches the data representation through feature engineering, so that subsequent time series analysis and leakage detection can be performed based on more comprehensive and meaningful features, significantly improving the leakage detection and prediction capabilities of the system. Through this series of preprocessing steps, the data preprocessing module provides a solid data foundation for the efficient operation of the entire leakage monitoring system.

[0099] Time Series Analysis Module

[0100] Regarding the time series analysis module, it should be noted that time series analysis plays a vital role in the urban pipeline leakage monitoring system. By analyzing and predicting the time series patterns of flow and pressure data, potential leakage trends and abnormal conditions can be discovered in advance, so as to prevent and respond quickly. The time series analysis module combines seasonality and trend decomposition with nonlinear time series analysis technology to provide accurate leakage prediction and diagnosis functions.

[0101] In this embodiment, the time series analysis module includes the following components:

[0102] Seasonality and trend decomposition unit: used to decompose time series data and identify periodic and long-term change patterns in the data. The specific operations are as follows:

[0103] First, the time series data is smoothed to eliminate the interference of short-term fluctuations on trends and seasonality.

[0104] The STL (Seasonal and Trend decomposition using Loess) method is used to decompose the time series data into three parts: trend, seasonality, and residual. The formula is as follows:

[0105] y t =T t +S t +R t

[0106] Among them, y t represents the observed value at time point t, T t is the trend component, S t is the seasonal component, R t is the remaining amount.

[0107] The Loess regression method is used to extract the trend component T t and seasonal component S t , the remaining part is taken as the residual R t , the residuals are further analyzed to identify abnormal fluctuations.

[0108] Nonlinear time series analysis unit: Use the nonlinear autoregressive moving average model (NARMA) to model and predict data and capture complex time-dependent structures. The specific operations are as follows:

[0109] Construct a NARMA model to predict future flow and pressure data. The model formula is as follows:

[0110]

[0111] Among them, c is a constant term, φ i is the autoregressive coefficient, θ j is the moving average coefficient, ∈ t is the error term at time point t, and p and q represent the orders of the autoregressive and moving average models, respectively.

[0112] Training model: Use historical data to train the model and determine the optimal model parameter φ i and θ j .

[0113] Model validation: The predictive performance of the model was evaluated through a cross-validation method to ensure that the model can accurately predict future leakage events.

[0114] Rolling forecast: Use the trained model to perform rolling forecasts to generate flow and pressure forecasts for a period of time in the future.

[0115] Anomaly detection: Based on the results of time series analysis, further identify abnormal fluctuations in the data to detect possible leakage events early. The specific operations are as follows:

[0116] Calculate statistical properties of residuals, such as mean and standard deviation, and set anomaly thresholds based on statistical principles.

[0117] When the residual exceeds the set threshold, it is marked as an abnormal data point and its possible cause is further analyzed.

[0118] In this embodiment, the time series analysis module not only provides a deep insight into historical data through comprehensive analysis and prediction of data, but also lays the foundation for future leakage detection and early warning. Through seasonality and trend decomposition, the system can identify periodicity and long-term trends in the data, thereby eliminating the interference of these factors on anomaly detection; through nonlinear time series analysis, the system can capture complex time-dependent structures and significantly improve the accuracy of leakage prediction.

[0119] For example, in a DMA partition, historical traffic data shows that the traffic change is the smallest between 2 a.m. and 4 a.m. every day. Through seasonal and trend decomposition, this pattern can be identified and incorporated into the model; and nonlinear time series analysis can further predict traffic changes in the next few hours. When the actual traffic data deviates significantly from the predicted value, the system can automatically trigger an alarm to indicate possible leakage.

[0120] Leakage detection module

[0121] Regarding the leakage detection module, it should be noted that this module is the core component of the urban pipeline leakage monitoring system. By comprehensively utilizing machine learning and statistical analysis techniques, the leakage detection module can accurately predict and detect leakage events in the water supply network, so that repair measures can be taken in time to reduce the waste of water resources. The design and implementation of this module greatly improves the leakage detection efficiency and accuracy of the system.

[0122] In this embodiment, the leakage detection module includes the following components:

[0123] Random forest regression model: used to analyze historical data and predict possible leakage events in the future. The specific operations are as follows:

[0124] Data preparation: Get cleaned and standardized data from the data preprocessing module, including features such as flow, pressure, and environmental variables.

[0125] Model training: Use historical data to train the random forest regression model and establish a prediction model for water flow and pressure. The random forest regression model improves the accuracy and robustness of the prediction by constructing multiple decision trees and combining the prediction results of these decision trees. The specific formula is as follows:

[0126]

[0127] in, is the predicted value, N is the number of decision trees, T i (x) is the prediction result of the i-th decision tree.

[0128] Feature importance analysis: By analyzing the importance of each feature in the model, the most critical features for leakage prediction are identified and the model is further optimized.

[0129] Density-based local anomaly factor (LOF) detection algorithm: used to detect abnormal conditions in the water supply network in real time, that is, potential leakage events. The specific operations are as follows:

[0130] Local reachable density calculation: First, the local reachable density (LRD) of each data point is calculated to measure the density of the data point in its neighborhood. The formula is as follows:

[0131]

[0132] Where x is the current data point, N k (x) is the set of k nearest neighbors of point x, reach-dist k (x,o) is the reachable distance between point x and its neighbor o.

[0133] Local anomaly factor calculation: Based on the local reachable density, the local anomaly factor (LOF) of each data point is calculated to identify the degree of anomaly relative to its neighboring points. The formula is as follows:

[0134]

[0135] Among them, lrd k (o) is the local reachability density of neighboring point o, lrd k (x) is the local reachable density of data point x. The larger the ratio is, the more likely point x is an outlier.

[0136] Abnormal detection and alarm: When an abnormality is detected, the system will automatically trigger an alarm and notify maintenance personnel to check and repair. The specific operations are as follows:

[0137] Abnormal threshold setting: According to the statistical characteristics of historical data, set a reasonable LOF threshold. When the LOF value of a data point exceeds the threshold, it is marked as abnormal.

[0138] Real-time monitoring and alarm: The system monitors flow and pressure data in real time. When an abnormality is detected, an alarm message is immediately generated and the relevant maintenance personnel are notified via SMS, email or other means.

[0139] Maintenance records and feedback: Each detected anomaly and processing result will be recorded to form a maintenance database for subsequent model optimization and maintenance strategy adjustment.

[0140] In this embodiment, the leakage detection module achieves efficient and accurate leakage prediction and detection by combining the random forest regression model and the LOF detection algorithm. The random forest regression model uses the water flow and pressure prediction model trained by historical data to effectively predict future water flow changes; the LOF detection algorithm detects abnormal conditions in real time by analyzing the density differences of data points in their local neighborhoods. The combination of the two enables the system to not only warn of potential leakage events in advance, but also respond quickly when leakage occurs, reducing water resource waste and economic losses.

[0141] For example, in a city's water supply network, the data collection module obtains flow and pressure data in real time. After being cleaned and standardized by the data preprocessing module, it is input into the random forest regression model for prediction. The model predicts the normal flow and pressure range for a period of time in the future. When the actual monitoring data deviates significantly from the predicted value, the LOF detection algorithm further analyzes the degree of abnormality of these data points and determines the possibility of leakage. If the LOF value exceeds the preset threshold, the system will immediately trigger an alarm and notify maintenance personnel to conduct on-site inspection and repair.

[0142] Noise analysis and positioning module

[0143] Regarding the noise analysis and positioning module, it should be noted that this module plays a key role in the urban pipeline leakage monitoring system. Through real-time monitoring and analysis of noise signals in the pipeline, the noise analysis and positioning module can accurately locate the location of the leak. This function greatly improves the accuracy of leakage detection and repair efficiency, reduces the waste of water resources and pipeline maintenance costs. Therefore, the design and implementation of the noise analysis and positioning module is crucial to the effective operation of the entire system.

[0144] In this embodiment, the noise analysis and positioning module includes the following components:

[0145] Noise feature extraction unit: used to convert the noise signal from the time domain to the frequency domain and extract the characteristic frequency components. The specific operations are as follows:

[0146] Noise signal acquisition: Use the hydrophone installed on the pipeline to capture the noise signal inside the pipeline in real time.

[0147] Time domain to frequency domain conversion: Fourier transform (FFT) is used to convert the time domain signal into the frequency domain signal. The formula is as follows:

[0148]

[0149] Where x(t) represents the noise signal strength at time t, f is the frequency, and X(f) is the complex frequency domain representation of the corresponding frequency f.

[0150] Extract characteristic frequencies: Analyze frequency domain signals and extract frequency components with specific water leakage noise characteristics, such as frequency peaks and bandwidth.

[0151] Noise pattern recognition unit: Use the improved K-means clustering algorithm to classify the noise data in the frequency domain to identify specific water leakage noise patterns. The specific operations are as follows:

[0152] Data preprocessing: The extracted characteristic frequency data are normalized to eliminate the differences under different measurement conditions.

[0153] Cluster analysis: K-means clustering algorithm is applied to classify the noise data into different noise patterns. The formula is as follows:

[0154]

[0155] Where k is the number of clusters, C i is the i-th cluster, μ i is the centroid of the ith cluster and x is a data point.

[0156] Leakage pattern recognition: Noise patterns with leakage characteristics are identified through clustering results and marked as potential leakage signals.

[0157] Leakage location unit: Use the noise data of multiple hydrophones to accurately locate the leakage point through triangulation. The specific operations are as follows:

[0158] Multi-point noise collection: Use multiple hydrophones distributed at different locations to collect noise data simultaneously and obtain the noise characteristics of multiple data points.

[0159] Triangulation method: Calculate the specific location of the leak based on the relative positions and noise characteristics of multiple hydrophones. The formula is as follows:

[0160]

[0161] Among them, (x i ,y i ) is the coordinate of the leakage point signal recorded by the ith hydrophone, n is the number of hydrophones, and Location represents the average coordinate of the leakage point calculated based on the collected noise data.

[0162] Verification of positioning results: By comparing the time difference and intensity of the noise signal, the accuracy of the calculated leak location can be further verified.

[0163] In this embodiment, the noise analysis and positioning module realizes the precise positioning of the pipeline leakage point through the above three steps. First, the noise feature extraction unit uses Fourier transform to convert the noise signal from the time domain to the frequency domain and extracts the key frequency components. Then, the noise pattern recognition unit classifies the noise data through the improved K-means clustering algorithm to identify the noise pattern with leakage characteristics. Finally, the leakage point positioning unit uses the data of multiple hydrophones to calculate the specific location of the leakage point through triangulation, and ensures the accuracy of the positioning result through multiple verifications.

[0164] For example, in the actual application of a city's water supply network, the noise signals collected in real time by the hydrophones installed on the pipelines indicate that a certain section of the pipeline may have a water leak. Through the noise feature extraction unit, the system converts these noise signals into frequency domain signals and extracts possible characteristic frequencies of water leaks. The noise pattern recognition unit then performs cluster analysis on these frequency data to identify possible water leak noise patterns. The leak location unit uses the noise data of multiple hydrophones to accurately calculate the location of the leak through triangulation, and finally verifies the existence of the leak on site and repairs it in a timely manner.

[0165] Real-time prediction alarm module

[0166] Regarding the real-time prediction and alarm module, it should be noted that this module plays a key role in the urban pipeline leakage monitoring system. Through real-time processing and analysis of data, the real-time prediction and alarm module can detect abnormal conditions in the pipeline network in a timely manner and quickly issue an alarm. This function can not only effectively reduce the waste of water resources caused by leakage, but also significantly reduce the risk of pipeline damage and maintenance costs. Therefore, the design and implementation of the real-time prediction and alarm module is crucial to ensure the efficient and safe operation of the water supply system.

[0167] In this embodiment, the real-time prediction alarm module includes the following components:

[0168] Real-time data processing unit: used to quickly process and analyze the real-time data transmitted from the data collection module. The specific operations are as follows:

[0169] Data reception: Receive data from various sensors in real time through the IoT gateway, including flow, pressure, noise, etc.

[0170] Data caching: Use efficient caching technology to temporarily store real-time data for fast access and processing.

[0171] Data filtering: Apply low-latency data filtering algorithms to remove obvious noise and outliers to ensure data quality for subsequent analysis.

[0172] Real-time prediction unit: Based on the pre-processed data, the operation status of the pipeline network is predicted in real time. The specific operations are as follows:

[0173] Short-term prediction model: Use the trained random forest regression model to make short-term predictions on real-time data and generate flow and pressure forecasts for the next few minutes to hours.

[0174] Dynamic update: Based on the latest data received, the parameters of the prediction model are continuously updated to ensure the accuracy and timeliness of the prediction results.

[0175] Alarm trigger unit: used to trigger an alarm when an abnormal situation is detected. The specific operations are as follows:

[0176] Anomaly Detection: Utilizes the Local Outlier Factor (LOF) algorithm to detect anomalies in data in real time.

[0177] Threshold setting: According to historical data and statistical analysis results, set reasonable alarm thresholds to ensure that anomalies can be detected in time without frequent false alarms.

[0178] Alarm mechanism: When an abnormality is detected, an alarm is triggered immediately and relevant personnel are notified through various means (such as SMS, email, application notification, etc.). The alarm information includes the details of the abnormal data and the possible leakage location for quick response and processing.

[0179] Alarm recording and feedback unit: used to record detailed information of each alarm and provide feedback mechanism. The specific operations are as follows:

[0180] Alarm records: Detailed records of each triggered alarm, including time, data status, prediction results and actual processing status.

[0181] Feedback mechanism: After handling an alarm event, maintenance personnel will provide feedback on the actual situation and handling results. The system will perform self-correction and optimization based on the feedback information to improve the accuracy of subsequent alarms.

[0182] In this embodiment, the real-time prediction alarm module realizes real-time monitoring and rapid response of the water supply network through the collaborative work of the above multiple units. First, the real-time data processing unit ensures the timeliness and accuracy of the data; second, the real-time prediction unit dynamically predicts the operating status of the network through the short-term prediction model; third, the alarm trigger unit quickly issues an alarm when an abnormality is detected and notifies relevant personnel; finally, the alarm recording and feedback unit records the detailed information of each alarm and continuously optimizes system performance through the feedback mechanism.

[0183] For example, in a city's water supply network, the real-time prediction alarm module receives real-time data from various sensors through the IoT gateway. After the data processing unit filters and caches the data, the real-time prediction unit uses the random forest regression model for short-term prediction. When the flow or pressure is detected to be significantly deviated from the normal range, the alarm trigger unit uses the LOF algorithm to further analyze the data anomaly and trigger the alarm. After receiving the alarm notification, the maintenance personnel quickly go to the site for inspection and processing, and at the same time feedback the actual situation and processing results to the system, and the system optimizes itself based on the feedback information.

[0184] The present invention realizes efficient and accurate monitoring and leakage detection of water supply pipe networks through the urban pipeline leakage monitoring system based on DMA zoning management technology. By integrating advanced data collection, preprocessing, time series analysis, leakage detection, noise analysis and positioning, and real-time prediction and alarm modules, the system can timely identify and locate leakage points, reduce water resource waste, reduce pipeline maintenance costs, and improve the operational efficiency and safety of the water supply system. In addition, the real-time prediction and alarm function significantly improves the response speed and accuracy of the system, ensuring that repair measures can be taken as soon as leakage occurs, thereby greatly improving the overall level of urban water supply management.

[0185] Please see attached Figure 3 The present invention also provides a method for monitoring urban pipeline leakage based on DMA zoning management technology, comprising the following steps:

[0186] Step S1: Data collection

[0187] In the first step, the system collects data from various sensing devices such as flow meters, pressure sensors and hydrophones in the DMA partition. These data include flow, pressure and noise signals, etc. Through these data, we can fully understand the operation status of the pipeline network and provide a basis for subsequent analysis.

[0188] Step S2: Data preprocessing

[0189] The collected data often contains noise and outliers, so in the second step, the system cleans, standardizes and performs feature engineering on the data. Through data cleaning, invalid data and outliers are removed; through standardization, data are analyzed at the same scale; through feature engineering, key features are extracted to enhance the data's expressiveness. This step ensures the quality and consistency of the data, laying the foundation for accurate leakage detection.

[0190] Step S3: Time series analysis

[0191] After data preprocessing is completed, the system performs seasonal and trend decomposition and nonlinear time series analysis on the preprocessed data. Seasonal and trend decomposition helps identify cyclical changes and long-term trends in the data, while nonlinear time series analysis is used to model and predict complex time-dependent structures. This step not only reveals the inherent laws of the data, but also provides a reliable prediction basis for leakage detection.

[0192] Step S4: Leakage detection

[0193] Based on the data processed by time series analysis, the system predicts and detects leakage through random forest regression model and density-based local anomaly factor detection algorithm. The random forest regression model is trained with historical data and can effectively predict future flow and pressure changes; the local anomaly factor detection algorithm identifies potential anomalies in real time by analyzing the local density differences of data points. This step achieves efficient detection and early warning of leakage events.

[0194] Step S5: Noise feature extraction and pattern recognition

[0195] In order to further accurately locate the leakage, the system uses the noise data of the hydrophone in the fifth step, converts the time domain signal into the frequency domain signal through Fourier transform, and uses the improved K-means clustering algorithm to extract noise features and pattern recognition. Through these steps, the system can identify noise signals with leakage characteristics, providing key data support for leak location.

[0196] Step S6: Leakage point location

[0197] After identifying the noise signal characteristic of the leak, the system uses triangulation to locate the leak. The specific location of the leak is calculated through the data collected by multiple hydrophones. The triangulation method uses the distance and signal time difference between each hydrophone to accurately calculate the coordinates of the leak. This step ensures the accurate location of the leak, thereby quickly guiding maintenance personnel to repair it.

[0198] Step S7: Real-time alarm and notification

[0199] When a leakage event is detected, the system processes data in real time and triggers an alarm to notify the relevant maintenance personnel. The alarm information can be sent to maintenance personnel through SMS, email, or APP to ensure that they can receive the notification and take action in time. Through this real-time alarm mechanism, the system can issue an early warning as soon as the leakage occurs, reducing the waste of water resources and the risk of pipeline damage.

[0200] Through the above steps, the urban pipeline leakage monitoring method of the present invention realizes comprehensive monitoring and leakage detection of the water supply network. The system can not only predict potential leakage events in advance, but also quickly locate and alarm when leakage occurs, greatly improving the operational efficiency and safety of the water supply system.

[0201] Please see attached Figure 4 The present invention further provides a computer device 40, comprising: a processor 41 and a memory 42, wherein the memory 42 stores a computer program executable by the processor, and when the computer program is executed by the processor, the above method is executed.

[0202] The present invention further provides a storage medium 43 on which a computer program is stored. When the computer program is run by the processor 41, the above method is executed.

[0203] Among them, the storage medium 43 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable red-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, disk or optical disk.

[0204] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An urban pipeline leakage monitoring system based on DMA zoning management technology, characterized in that: include: Data collection module, used to collect data from flow meters, pressure sensors and hydrophones in DMA zones; A data preprocessing module, used for preprocessing the data; Time series analysis module, used to perform seasonality and trend decomposition and nonlinear time series analysis on preprocessed data; The leakage detection module includes a random forest regression model and a density-based local anomaly factor detection algorithm for predicting and detecting leakage; The noise analysis and positioning module is used to determine the specific location of the pipe leak through the noise data analysis of the hydrophone; Real-time prediction and alarm module, which processes data in real time and triggers alarms when leakage is detected.

2. According to claim 1, a city pipeline leakage monitoring system based on DMA zoning management technology is characterized in that: The data collection module is used to collect data from various sensor devices, and the sensors include: Flow sensors are located at the inlet and outlet of each DMA partition to measure the water flow through a specific pipe section to monitor the water flow dynamics in the area. The data types include instantaneous flow and accumulated flow. Pressure sensors are installed at key pipe sections, including zone inlets, outlets and potential high-risk areas, to monitor water pressure changes in real time and facilitate early identification of abnormal pressure fluctuations in the system; Hydrophones are deployed in pipe sections near known and potential leaks to capture sound signals caused by water flow for subsequent leak location analysis; Auxiliary sensing equipment, including but not limited to temperature sensors and water quality monitoring equipment, is used to provide environmental and water quality data to support the analysis of environmental factors for leakage detection.

3. According to the urban pipeline leakage monitoring system based on DMA zoning management technology as described in claim 1, it is characterized in that: The data preprocessing module comprises: Data cleaning unit to remove data during non-working hours, correct obvious erroneous readings, and fill in missing values; Data normalization unit, used to normalize data from different sources and scopes to a uniform scale; The feature engineering unit is used to extract key features from the raw data that are helpful for leakage detection and prediction, including extracting statistical features, time features, and frequency domain features from flow and pressure time series data.

4. According to claim 1, a city pipeline leakage monitoring system based on DMA zoning management technology is characterized in that: The time series analysis module includes: The seasonality and trend decomposition unit is used to perform seasonality and trend decomposition on time series data to identify cyclical and long-term change patterns in the data. The decomposition is performed using the following formula: y t =T t +S t +R t Among them, y t represents the observed value at time point t, T t is the trend component, S t is the seasonal component, R t is the remaining amount; The nonlinear time series analysis unit uses a nonlinear autoregressive moving average model to simulate and predict time series data using the following model formula: Among them, c is a constant term, φ i is the autoregressive coefficient, θ j is the moving average coefficient, ∈ t is the error term at time point t, and p and q represent the orders of the autoregressive and moving average models, respectively.

5. The urban pipeline leakage monitoring system based on DMA zoning management technology according to claim 1 is characterized in that: The leakage detection module includes a random forest regression model for predicting leakage probability and impact range. The model is trained with historical data and calculates the importance of each feature.

6. The urban pipeline leakage monitoring system based on DMA zoning management technology according to claim 1 is characterized in that: The local abnormal factor detection algorithm of the leakage detection module includes: The local reachability density is calculated for each observation point to measure the local density of the observation point. The formula is as follows: Where x is the current data point, N k (x) is the set of k nearest neighbors of point x, reach-dist k (x, o) is the reachable distance between point x and its neighbor o; Based on the above local reachable density, the local anomaly factor is calculated for each data point to identify the degree of anomaly relative to its neighboring points. The formula is as follows: Among them, lrd k (o) is the local reachability density of neighboring point o, lrd k (x) is the local reachable density of data point x. The larger the ratio is, the more likely point x is an outlier.

7. The urban pipeline leakage monitoring system based on DMA zoning management technology according to claim 1 is characterized in that: The noise analysis and positioning module includes: The noise feature extraction unit uses Fourier transform to convert the noise signal in the time domain to the frequency domain for analysis. The formula is as follows: Where x(t) represents the noise signal strength at time t, f is the frequency, and X(f) is the complex frequency domain representation of the corresponding frequency f; The noise pattern recognition unit uses an improved K-means clustering algorithm to classify the noise data in the frequency domain to identify specific water leakage noise patterns. First, the distance from each noise sample to each class center is calculated, and then the class center is updated. The iteration is repeated until the convergence condition is met; The leak location unit uses the noise data collected by multiple hydrophones installed at different locations to locate the leak through triangulation, using the following formula: Among them, (x i ,y i ) is the coordinate of the leakage point signal recorded by the ith hydrophone, n is the number of hydrophones, and Location represents the average coordinate of the leakage point calculated based on the collected noise data.

8. The urban pipeline leakage monitoring system based on DMA zoning management technology according to claim 1 is characterized in that: The real-time prediction alarm module includes: Real-time data processing unit, which uses stream processing framework to clean, process and analyze data; The alarm system triggers an alarm when the predicted leakage probability or LOF score exceeds the threshold, and notifies maintenance personnel via SMS, email or APP.

9. A method for monitoring urban pipeline leakage based on DMA zoning management technology, based on the system according to any one of claims 1 to 8, characterized in that: The following steps are involved: Collect data from flow meters, pressure sensors and hydrophones within the DMA zone; Cleaning, standardizing and feature engineering the data; Perform seasonality and trend decomposition and nonlinear time series analysis on preprocessed data; Predict and detect leakage through random forest regression model and density-based local anomaly factor detection algorithm; The noise data from the hydrophone is used to extract noise features and perform pattern recognition through Fourier transform and improved K-means clustering algorithm. Use triangulation to locate leaks; Process data in real time and trigger an alarm when a leak is detected, notifying maintenance personnel via SMS, email or APP.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to claim 9 is implemented.

Images