Water supply network leakage detection method and device

Abstract

The invention provides a water supply pipe network leakage detection method and device, and the method comprises the steps: obtaining a pressure data set in response to a pipe network monitoring trigger signal, and carrying out the segmentation processing of the pressure data set, and obtaining a pressure data set and staged pressure data containing a historical pressure data set; performing associated feature extraction on the data in the pressure data group set, and generating a pressure feature data set through grouping, redundancy filtering, matching degree evaluation and aggregation processing; retrieving the data set based on the staged pressure data to obtain a target pressure data set; and performing updating processing on the target data set by using the historical pressure data set, and generating a target leakage feature sequence as a detection result through associated feature extraction, score calculation and sorting. According to the method, the accuracy and efficiency of pipe network leakage detection can be improved, the leakage area can be quickly positioned, and water supply pipe network management can be optimized.

Description

Technical Field

[0001] The present invention relates to the technical field of water supply network monitoring, and more specifically, to a method and device for detecting leakage of a water supply network. Background Art

[0002] In today's society, with the acceleration of urbanization and the increasing demand for water resources, the water supply network system plays a vital role in urban infrastructure. However, the problem of water supply network leakage has always been one of the major challenges facing the water supply industry. Traditional methods for detecting water supply network leakage mainly rely on manual inspections, pressure monitoring, and flow analysis. Manual inspections require a lot of manpower and material resources, and the detection efficiency is low, making it difficult to detect small leaks in a timely manner; while traditional pressure monitoring and flow analysis methods can provide certain leakage information, but they can often only be detected when the leakage is more serious, and are easily affected by factors such as pipe network pressure fluctuations and changes in water use patterns, resulting in insufficient accuracy and reliability of leakage detection.

[0003] Among the existing leakage detection technologies, some methods install pressure sensors and flow sensors in the pipeline network to monitor the pressure and flow changes of the pipeline network in real time, and then use data processing algorithms to analyze these data to determine whether there is a leakage. Although these methods have improved the automation of leakage detection to a certain extent, they still have some limitations. For example, the installation and maintenance costs of sensors are high, and the accuracy and reliability of sensors are affected by environmental factors. In addition, data processing algorithms usually require a large amount of historical data for training and optimization, but in practical applications, it is often difficult to obtain sufficiently accurate and comprehensive historical data. Therefore, these methods are difficult to achieve efficient and accurate leakage detection in practical applications.

[0004] In the process of implementing the embodiments of the present invention, the inventors found that there are at least the following problems or defects in the prior art: the existing leakage detection method cannot effectively respond to the leakage detection needs under complex working conditions of the pipeline network, and it is difficult to timely detect leakage and accurately locate the leakage position at the early stage of leakage; at the same time, the existing method is highly dependent on sensor data and is easily affected by sensor accuracy and environmental interference, resulting in insufficient accuracy and reliability of leakage detection. In addition, when processing large-scale pipeline network data, the existing technology has a high computational complexity and is difficult to meet real-time requirements. Summary of the invention

[0005] The invention provides a method and device for detecting leakage of a water supply network.

[0006] In a first aspect of the present invention, a method for detecting leakage in a water supply network is provided, comprising: In response to receiving a pipeline network monitoring trigger signal, acquiring a pressure data set; Segmenting the pressure data set to obtain a pressure data group set and staged pressure data, wherein the staged pressure data includes a historical pressure data set; Extracting associated features from each pressure data in the pressure data set to obtain a pressure feature data set; Based on the staged pressure data, the pressure characteristic data set is retrieved and processed to obtain a target pressure data set; Based on the historical pressure data set, the target pressure data set is updated to obtain a target leakage feature sequence for output as a leakage detection result.

[0007] Furthermore, the extracting of associated features from each pressure data in the pressure data set to obtain a pressure feature data set includes: performing grouping processing on each pressure data in the pressure data group set to obtain an associated pressure data group set; performing redundancy filtering on each associated pressure data group in the associated pressure data group set to obtain a target associated pressure data group set; Performing matching evaluation on each target-associated pressure data group in the target-associated pressure data group set to obtain a pressure-to-pressure matching data set; Aggregation processing is performed on the pressure matching degree dataset to obtain a pressure feature dataset.

[0008] Furthermore, the performing matching evaluation on each target-associated pressure data set in the target-associated pressure data set to obtain a pressure matching data set includes: For each target-associated pressure data group in the target-associated pressure data group set, performing the following steps: generating an inter-pressure association data set based on the pressure data group set and the target-associated pressure data group; Based on the pressure-related data set, a matching evaluation is performed on the target-related pressure data set to obtain pressure-related matching data.

[0009] Further, generating the pressure-related data set based on the pressure data set set and the target-related pressure data set includes: Determine a piece of target-associated pressure data in the target-associated pressure data group that satisfies a first preset index condition as first pressure data; Determine a piece of target-associated pressure data in the target-associated pressure data group that satisfies a second preset index condition as second pressure data; For each pressure data group in the pressure data group set, the following steps are performed: determining a pressure data in the pressure data group that matches the first pressure data as target first pressure data; determining a piece of pressure data in the pressure data group that matches the second pressure data as target second pressure data; A matching degree evaluation is performed on the target first pressure data and the target second pressure data to obtain pressure correlation data.

[0010] Furthermore, the updating process of the target pressure data set based on the historical pressure data set to obtain a target leakage feature sequence includes: For each target pressure data in the target pressure data set, the following steps are performed: extracting correlation features between the historical pressure data set and the target pressure data to obtain a set of pressure matching degree adjustment values; Determining the sum of the respective pressure matching degree adjustment amounts in the pressure matching degree adjustment amount set as the target pressure score; Determine the leakage characteristic data corresponding to the target pressure data and the target pressure score as target leakage characteristic data; The determined target leakage characteristic data are sorted to obtain a target leakage characteristic sequence.

[0011] Further, each pressure data in the pressure data set includes a pipe network area identification group; and the extraction of associated features of the historical pressure data group and the target pressure data to obtain a pressure matching adjustment amount set includes: For each historical pressure data in the historical pressure data set, the following steps are performed: determining one pressure data in the pressure data set that matches the target pressure data as user pressure data; Performing a matching evaluation on the user pressure data and the historical pressure data to obtain a historical matching value; Performing a search process on the pressure feature data set to obtain a target initial matching value; Determining the historical pressure data, the historical matching value, and the target initial matching value as matching adjustment data; Based on the matching degree adjustment data, an adjustment amount of matching degree between pressures is generated.

[0012] Furthermore, the performing matching evaluation on the user pressure data and the historical pressure data to obtain a historical matching value includes: From the pipe network area identification group included in the user pressure data, a pipe network area identification that matches the pipe network area identification group included in the historical pressure data is selected as a public area identification to obtain a public area identification group; Performing feature embedding processing on the user pressure data and the historical pressure data to obtain a target pressure feature vector and a historical pressure feature vector; A historical matching value is generated based on the common area identification group, the target pressure feature vector and the historical pressure feature vector.

[0013] In a second aspect of the present invention, a water supply network leakage detection device is provided, comprising: an acquisition unit, configured to acquire a pressure data set in response to receiving a pipeline network monitoring trigger signal; a segmentation processing unit configured to perform segmentation processing on the pressure data set to obtain a pressure data group set and staged pressure data, wherein the staged pressure data includes a historical pressure data group; a correlation feature extraction unit configured to extract correlation features from each pressure data in the pressure data set to obtain a pressure feature data set; A retrieval processing unit is configured to perform retrieval processing on the pressure characteristic data set based on the staged pressure data to obtain a target pressure data set; The update processing unit is configured to update the target pressure data set based on the historical pressure data set to obtain a target leakage feature sequence for output as a leakage detection result.

[0014] In a third aspect of the present invention, an electronic device is provided, comprising: at least one processor, a memory and an input-output unit; wherein the memory is used to store a computer program, and the processor is used to call the computer program stored in the memory to execute any one of the methods described in the first aspect.

[0015] In a fourth aspect of the present invention, a computer-readable storage medium is provided, which includes instructions, and when the instructions are executed on a computer, the computer executes any one of the methods in the first aspect.

[0016] The above-mentioned embodiments of the present invention have at least the following beneficial effects: the method can construct a pressure feature data set based on multi-stage correlation analysis of pipeline network pressure data through grouping processing, redundant filtering and matching evaluation, and dynamically update it in combination with historical pressure data, thereby improving the accuracy of leakage detection. Through the calculation of the matching adjustment amount between pressures and the evaluation of the target pressure score, a more reliable target leakage feature sequence can be generated, effectively distinguishing normal pressure fluctuations from real leakage signals.

[0017] This technology can optimize the monitoring efficiency of water supply network, reduce data processing redundancy and improve calculation speed through phased pressure data retrieval and feature aggregation processing. Combined with pipe network area identification matching and feature embedding technology, it can enhance the accuracy of leakage location, provide more reliable decision-making basis for water supply system maintenance, and reduce the false alarm rate and missed alarm rate of leakage detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood by reading the detailed description below with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, in which: Figure 1 A schematic diagram of a flow chart of a water supply network leakage detection method provided by an embodiment of the present invention; Figure 2 A schematic diagram of the structure of a water supply network leakage detection device provided by an embodiment of the present invention; Figure 3 The schematic diagram schematically shows the structure of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0019] The principles and spirit of the present invention will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided only to enable those skilled in the art to better understand and implement the present invention, and are not intended to limit the scope of the present invention in any way. On the contrary, these embodiments are provided to make the present invention more thorough and complete, and to fully convey the scope of the present invention to those skilled in the art.

[0020] Those skilled in the art will appreciate that the embodiments of the present invention may be implemented as a device, apparatus, equipment, method or computer program product. Therefore, the present invention may be implemented in the following forms, namely: complete hardware, complete software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.

[0021] It should be noted that any number of elements in the drawings is for illustration rather than limitation, and any naming is only for distinction and does not have any limiting meaning.

[0022] Reference below Figure 1 , Figure 1 The following is a flow chart of a water supply network leakage detection method provided by an embodiment of the present invention. Figure 1 As shown, a water supply network leakage detection method includes: S1. In response to receiving a pipeline network monitoring trigger signal, obtaining a pressure data set; S2. Segmenting the pressure data set to obtain a pressure data set and staged pressure data, wherein the staged pressure data includes a historical pressure data set; S3. Extracting associated features from each pressure data in the pressure data set to obtain a pressure feature data set; S4. Based on the staged pressure data, the pressure feature data set is retrieved and processed to obtain a target pressure data set; S5. Based on the historical pressure data set, the target pressure data set is updated to obtain a target leakage feature sequence for output as a leakage detection result.

[0023] It should be noted that in the embodiment of the present invention, a pressure data set is first obtained in response to receiving a pipeline monitoring trigger signal. The pipeline monitoring trigger signal here refers to a signal used to start the leakage detection process, which can be a timing trigger signal or a signal generated according to the operating status of the pipeline network (such as abnormal pressure fluctuations). The pressure data set refers to a set of pressure data collected from multiple monitoring points in the water supply pipeline network. These data reflect the pressure conditions of the pipeline network at different locations and times, and are the basic information for leakage detection. By obtaining these data, the original basis can be provided for subsequent leakage detection.

[0024] Specifically, each pressure data in the pressure data set includes a pipe network area identification group, which is used to identify the pipe network area to which the data belongs, so that the pressure conditions in different areas can be distinguished in subsequent processing. The pressure data set can be collected by pressure sensors installed in the pipe network. These sensors are distributed at key nodes of the pipe network, such as pump station outlets, branch pipelines, etc. The collected pressure data will be uploaded to the data processing system at a certain frequency (such as once per minute) to form a pressure data set. In addition, each pressure data in the pressure data set may also contain a timestamp to record the specific time of data collection for time series analysis.

[0025] Preferably, the process of acquiring the pressure data set can be further refined. For example, in order to improve the accuracy and completeness of the data, the sensor can be calibrated during the data collection stage and a reasonable collection frequency can be set. The collection frequency can be adjusted according to the scale and operating characteristics of the pipeline network. For large and complex pipeline networks, the collection frequency can be appropriately increased. At the same time, in order to deal with possible sensor failures or data anomalies, a data quality check mechanism can be set up to perform preliminary screening of the collected pressure data and remove obviously erroneous or abnormal data. In addition, in order to facilitate subsequent processing, the collected pressure data can be stored and organized according to the pipeline network area and time sequence to form a structured pressure data set.

[0026] In some embodiments, extracting associated features from each pressure data in the pressure data set to obtain a pressure feature data set includes: performing grouping processing on each pressure data in the pressure data group set to obtain an associated pressure data group set; performing redundancy filtering on each associated pressure data group in the associated pressure data group set to obtain a target associated pressure data group set; Performing matching evaluation on each target-associated pressure data group in the target-associated pressure data group set to obtain a pressure-to-pressure matching data set; Aggregation processing is performed on the pressure matching degree dataset to obtain a pressure feature dataset.

[0027] It should be noted that the step of extracting associated features from each pressure data in the pressure data group set to obtain a pressure feature data set in the embodiment of the present invention is a key link in the leakage detection method. This process aims to extract feature information related to leakage from a large amount of pressure data so that leakage can be identified more accurately later. Associated feature extraction refers to extracting features that can reflect changes in the state of the pipeline network by analyzing the relationships and patterns between pressure data. The pressure data group set refers to a set of pressure data grouped according to certain rules (such as time, area, etc.), and the pressure feature data set is a set of extracted feature information, which will serve as an important basis for subsequent leakage detection.

[0028] Specifically, each pressure data in the pressure data set first needs to be grouped and processed to obtain an associated pressure data set. Grouping processing here refers to dividing the pressure data into multiple associated pressure data groups according to information such as the topological structure of the pipeline network, the positional relationship of the monitoring points or the time series. Each associated pressure data group contains a set of interrelated pressure data, for example, pressure data at different monitoring points at the same time point, or pressure data at different time points at the same monitoring point. Next, each associated pressure data group in the associated pressure data set is redundantly filtered to obtain a target associated pressure data set. The purpose of redundant filtering is to remove duplicate or highly similar data to reduce the amount of data and improve the efficiency of subsequent processing. Finally, each target associated pressure data group in the target associated pressure data set is matched and evaluated to obtain a pressure matching dataset. Matching evaluation is achieved by calculating the similarity or correlation between pressure data. For example, correlation coefficients, Euclidean distances and other methods can be used to evaluate the matching between pressure data.

[0029] Preferably, the process of extracting associated features can be further refined in the following ways. First, during grouping, the pressure data can be grouped by region based on the regional division of the pipeline network and the distribution of monitoring points. At the same time, combined with the time series information, the pressure data in the same region can be arranged in chronological order to form a time series grouping. Secondly, during the redundant filtering process, a threshold can be set to judge the similarity between the data. When the similarity between two pressure data is higher than the set threshold, they are considered to be redundant data and only one of them is retained. In addition, when evaluating the matching degree, a combination of multiple methods can be used. For example, the correlation coefficient between the pressure data is first calculated, and then a comprehensive evaluation is performed in combination with the Euclidean distance to improve the accuracy of the matching degree evaluation. Finally, the evaluated pressure matching data is aggregated to form a pressure feature data set, which provides richer feature information for subsequent leakage detection.

[0030] In some embodiments, performing matching evaluation on each target-associated pressure data set in the target-associated pressure data set to obtain a pressure matching dataset includes: For each target-associated pressure data group in the target-associated pressure data group set, performing the following steps: generating an inter-pressure association data set based on the pressure data group set and the target-associated pressure data group; Based on the pressure-related data set, a matching evaluation is performed on the target-related pressure data set to obtain pressure-related matching data.

[0031] It should be noted that the step of evaluating the matching degree of each target-associated pressure data group in the target-associated pressure data group set to obtain the pressure matching degree data set in the embodiment of the present invention is a key link in the process of extracting the associated features. This process aims to quantify the matching degree between target-associated pressure data groups by analyzing the similarity or correlation between them, thereby providing basic data for subsequent leakage feature extraction. The pressure matching degree data set is a data set generated by evaluating the matching degree between target-associated pressure data groups. It reflects the strength of association between different pressure data and is an important basis for subsequent leakage detection.

[0032] Specifically, the target-associated pressure data set refers to a set of pressure data sets after redundancy filtering, wherein each target-associated pressure data set contains a set of mutually related pressure data. The matching evaluation process is to generate a pressure-related data set based on the pressure data set set and the target-associated pressure data set, and then evaluate the matching degree based on these related data. The pressure-related data set refers to a data set generated by analyzing the relationship between the target-associated pressure data set and other pressure data, and it contains the association information between the pressure data. When evaluating the matching degree, it can be achieved by calculating the similarity indicators between the pressure data (such as correlation coefficient, Euclidean distance, etc.), which can quantify the degree of association between the pressure data.

[0033] Preferably, the matching evaluation process can be further refined in the following ways. First, when generating a pressure-related data set, a feature vector extraction method can be used to convert the pressure data in each target-related pressure data group into a feature vector, and then the similarity between these feature vectors is calculated. For example, the time series characteristics, frequency characteristics, etc. of the pressure data can be extracted as components of the feature vector. Secondly, when evaluating the matching degree, a comprehensive evaluation can be performed in combination with multiple similarity indicators. For example, the correlation coefficient between the pressure data is first calculated to evaluate their linear correlation; then the Euclidean distance is calculated to evaluate their numerical differences. By combining these two indicators, the matching degree between the pressure data can be more comprehensively evaluated. In addition, a weight mechanism can be introduced to assign weights to similarity indicators according to the importance of different features, so as to further improve the accuracy and reliability of the matching evaluation. Finally, the evaluated pressure matching data are aggregated to form a pressure matching data set, which provides more accurate input data for subsequent leakage feature extraction.

[0034] In some embodiments, generating the pressure correlation data set based on the pressure data set set and the target-related pressure data set includes: Determine a piece of target-associated pressure data in the target-associated pressure data group that satisfies a first preset index condition as first pressure data; Determine a piece of target-associated pressure data in the target-associated pressure data group that satisfies a second preset index condition as second pressure data; For each pressure data group in the pressure data group set, the following steps are performed: determining a pressure data in the pressure data group that matches the first pressure data as target first pressure data; determining a piece of pressure data in the pressure data group that matches the second pressure data as target second pressure data; A matching degree evaluation is performed on the target first pressure data and the target second pressure data to obtain pressure correlation data.

[0035] It should be noted that the process of processing each pressure data group in the pressure data group set to generate pressure correlation data in the embodiment of the present invention is an important basis for matching evaluation. This process determines the specific pressure data (first pressure data and second pressure data) in the target associated pressure data group, and searches for the matching pressure data (target first pressure data and target second pressure data) in the pressure data group set, and then evaluates the matching degree between them, and finally generates pressure correlation data. This method can effectively capture the correlation between pressure data and provide a more accurate basis for subsequent leakage detection.

[0036] Specifically, the first preset index condition and the second preset index condition are rules for selecting specific pressure data from the target associated pressure data group. For example, the first preset index condition may be to select the pressure data with the highest or lowest pressure value as the first pressure data, and the second preset index condition may be to select the pressure data with a certain time interval with the first pressure data as the second pressure data. The target associated pressure data group is extracted from the set of pressure data groups after redundant filtering, and is a subset for further analysis. Each pressure data group in the set of pressure data groups contains multiple pressure data, which may come from different times of the same monitoring point, or the same time of different monitoring points. During the processing, for each pressure data group, it is necessary to find pressure data that matches the first pressure data and the second pressure data in the target associated pressure data group, that is, the target first pressure data and the target second pressure data. The matching here may mean that the pressure values ​​are the same within a certain error range, or the timestamps are the same, etc. Then, the matching degree of the target first pressure data and the target second pressure data is evaluated to obtain the pressure association data, which will be used for subsequent aggregation processing.

[0037] Preferably, the process of generating pressure correlation data can be further refined in the following ways. First, the preset index conditions can be flexibly set according to the actual application scenario. For example, the data with the largest pressure change rate can be selected as the first pressure data according to the operating characteristics of the pipeline network to capture the drastic fluctuations of the pipeline network pressure; or the data with a pressure value within a certain range can be selected as the second pressure data to focus on the normal operating pressure interval of the pipeline network. Secondly, in the matching process, a fault-tolerant mechanism can be introduced to allow a certain error range to improve the flexibility and accuracy of the matching. For example, the error range of the pressure value is set to ±0.1MPa, and the error range of the timestamp is ±1 minute. In addition, the matching evaluation can adopt a more complex algorithm, such as a similarity evaluation model based on machine learning. The model can learn the complex relationship between the pressure data through training, and output the matching score between them according to the input target first pressure data and the target second pressure data. The input parameters of the model may include pressure value, timestamp, pipeline area identification, etc. Through the combination of these parameters, the model can more comprehensively evaluate the correlation between the pressure data. Finally, the evaluated pressure correlation data are aggregated to form a pressure matching dataset, providing richer and more accurate feature information for subsequent leakage feature extraction.

[0038] In some embodiments, the updating process of the target pressure data set based on the historical pressure data set to obtain a target leakage feature sequence includes: For each target pressure data in the target pressure data set, the following steps are performed: extracting correlation features between the historical pressure data set and the target pressure data to obtain a set of pressure matching degree adjustment values; Determining the sum of the respective pressure matching degree adjustment amounts in the pressure matching degree adjustment amount set as the target pressure score; Determine the leakage characteristic data corresponding to the target pressure data and the target pressure score as target leakage characteristic data; The determined target leakage characteristic data are sorted to obtain a target leakage characteristic sequence.

[0039] It should be noted that the process of updating the target pressure data set based on the historical pressure data set to obtain the target leakage feature sequence in the embodiment of the present invention is a key step in the leakage detection method. This process extracts data that can reflect the leakage characteristics by analyzing the correlation characteristics between the target pressure data and the historical pressure data, and sorts them to generate a target leakage feature sequence, thereby providing a clear basis for leakage detection. The historical pressure data set refers to a set of pressure data collected over a period of time in the past. It contains the pressure characteristics under normal operating conditions and can be used for comparative analysis with the current target pressure data. The target leakage feature sequence is an ordered data sequence obtained by updating the target pressure data set, in which the leakage feature data can effectively reflect the possibility and severity of the leakage.

[0040] Specifically, for each target pressure data in the target pressure data set, it is necessary to extract associated features with the historical pressure data group to obtain a pressure matching adjustment set. In this process, the pressure matching adjustment set is generated by evaluating the similarity or correlation change between the target pressure data and the historical pressure data. The leakage feature data and the target pressure score corresponding to each target pressure data together constitute the target leakage feature data. The target pressure score is obtained by summing the pressure matching adjustments in the pressure matching adjustment set, which reflects the overall matching degree between the target pressure data and the historical data. Finally, the target leakage feature sequence is generated by sorting the target leakage feature data. The basis for sorting can be the size of the target pressure score. The lower the score, the greater the possibility of leakage.

[0041] Preferably, the updating process can be further refined in the following ways. First, when extracting the associated features, the feature embedding technology can be used to convert the target pressure data and the historical pressure data into feature vectors, and then the pressure matching adjustment amount is generated by calculating the similarity between the feature vectors. For example, the matching degree between the feature vectors can be evaluated using methods such as cosine similarity or Euclidean distance. Secondly, the calculation of the target pressure score can introduce a weight mechanism to assign weights to the matching adjustment amount according to the importance of different features, so as to more accurately reflect the leakage characteristics of the target pressure data. For example, a higher weight can be given to the historical pressure data related to the key area of ​​the pipeline network. In addition, the sorting process can be adjusted according to the actual needs of leakage detection. For example, in addition to sorting according to the target pressure score, other factors (such as the timestamp of the pressure data, the pipeline area identification, etc.) can also be combined for comprehensive sorting to improve the accuracy and reliability of leakage detection. The target leakage feature sequence finally generated can be directly used for leakage detection or as basic data for further analysis.

[0042] In some embodiments, each pressure data in the pressure data set includes a pipe network area identification group; and the extraction of associated features of the historical pressure data group and the target pressure data to obtain a pressure matching adjustment amount set includes: For each historical pressure data in the historical pressure data set, the following steps are performed: determining one pressure data in the pressure data set that matches the target pressure data as user pressure data; Performing a matching evaluation on the user pressure data and the historical pressure data to obtain a historical matching value; Performing a search process on the pressure feature data set to obtain a target initial matching value; Determining the historical pressure data, the historical matching value, and the target initial matching value as matching adjustment data; Based on the matching degree adjustment data, an adjustment amount of matching degree between pressures is generated.

[0043] It should be noted that the process of evaluating the matching degree of user pressure data and historical pressure data and obtaining the historical matching degree value in the embodiment of the present invention is an important step in the extraction of leakage characteristics. This process generates a historical matching degree value by analyzing the similarity between user pressure data and historical pressure data, thereby providing a basis for the generation of subsequent pressure matching adjustment amounts. User pressure data refers to pressure data that matches the target pressure data, which reflects the pressure characteristics under the current state of the pipeline network; and the historical matching degree value is a quantitative indicator obtained by evaluating the similarity between user pressure data and historical pressure data, which is used to measure the degree of matching between the two.

[0044] Specifically, the user pressure data is extracted from the pressure data set and is the pressure data with the same pipe network area identification group as the target pressure data. The pipe network area identification group is used to identify the pipe network area to which the pressure data belongs, and is an important attribute of the pressure data. When performing the matching evaluation, it is first necessary to select the pipe network area identification that matches the pipe network area identification group of the historical pressure data from the pipe network area identification group of the user pressure data to form a common area identification group. This process ensures the regional correlation between the user pressure data and the historical pressure data. Then, the user pressure data and the historical pressure data are subjected to feature embedding processing to generate a target pressure feature vector and a historical pressure feature vector. Feature embedding processing is a process of converting raw data into feature vectors, which can better reflect the intrinsic characteristics of the data. Finally, based on the common area identification group, the target pressure feature vector and the historical pressure feature vector, a historical matching value is generated. The generation of the historical matching value can be achieved by calculating the similarity between the feature vectors, for example, using methods such as cosine similarity or Euclidean distance.

[0045] Preferably, the matching evaluation process can be further refined in the following ways. First, in the feature embedding process, a deep learning model, such as a convolutional neural network (CNN) or a recurrent neural network (RNN), can be used to convert the user pressure data and the historical pressure data into feature vectors. These models can automatically learn complex features in the data and generate feature vectors with high expressiveness. For example, CNN can be used to extract local features in pressure data, while RNN can be used to process the time series features of pressure data. Secondly, when calculating the historical matching value, a weighting mechanism can be introduced to assign weights to each dimension of the feature vector according to the importance of different features. For example, a higher weight can be given to feature dimensions that are highly correlated with leakage detection. In addition, a comprehensive evaluation can be combined with other features (such as the timestamp of the pressure data, the topological structure of the pipe network area, etc.) to improve the accuracy and reliability of the historical matching value. Finally, based on the generated historical matching value, the pressure-to-pressure matching adjustment amount can be further generated to provide a more accurate basis for leakage detection.

[0046] In some embodiments, performing a matching evaluation on the user pressure data and the historical pressure data to obtain a historical matching value includes: From the pipe network area identification group included in the user pressure data, a pipe network area identification that matches the pipe network area identification group included in the historical pressure data is selected as a public area identification to obtain a public area identification group; Performing feature embedding processing on the user pressure data and the historical pressure data to obtain a target pressure feature vector and a historical pressure feature vector; A historical matching value is generated based on the common area identification group, the target pressure feature vector and the historical pressure feature vector.

[0047] It should be noted that in the embodiment of the present invention, in the process of performing matching evaluation on user pressure data and historical pressure data to obtain historical matching values, the specific steps of matching evaluation are further refined. This process not only takes into account the numerical characteristics of the pressure data, but also combines the matching of the pipe network area identification group, thereby more comprehensively reflecting the similarity between the user pressure data and the historical pressure data. In this way, the matching degree between the pressure data can be more accurately evaluated, thereby providing a more reliable basis for leakage detection. This method is particularly suitable for extracting leakage features in complex pipe network systems because it can comprehensively consider the topological structure of the pipe network and the dynamic changes of pressure data.

[0048] Specifically, user pressure data is extracted from the pressure data set and is pressure data with the same pipe network area identification group as the target pressure data, which reflects the pressure characteristics under the current pipe network state. Historical pressure data is pressure data collected in the past period of time, which is used for comparative analysis with user pressure data. The pipe network area identification group is an important attribute of pressure data, which is used to identify the pipe network area to which the pressure data belongs. In the matching evaluation process, firstly, the part matching the pipe network area identification group of the historical pressure data is selected from the pipe network area identification group of the user pressure data to form a common area identification group. This process ensures the regional correlation between the user pressure data and the historical pressure data. Then, feature embedding processing is performed on the user pressure data and the historical pressure data to generate a target pressure feature vector and a historical pressure feature vector. Feature embedding processing is a process of converting raw data into feature vectors, which can better reflect the intrinsic characteristics of the data. Finally, based on the common area identification group, the target pressure feature vector and the historical pressure feature vector, a historical matching value is generated. The generation of the historical matching value can be achieved by calculating the similarity between the feature vectors, for example, using methods such as cosine similarity or Euclidean distance. This method not only considers the numerical characteristics of pressure data, but also combines the matching of pipeline area identification groups, thereby more comprehensively reflecting the similarity between user pressure data and historical pressure data.

[0049] Preferably, the matching evaluation process can be further refined in the following ways. First, in the feature embedding process, a deep learning model, such as a convolutional neural network (CNN) or a recurrent neural network (RNN), can be used to convert the user pressure data and the historical pressure data into feature vectors. These models can automatically learn complex features in the data and generate feature vectors with high expressiveness. For example, CNN can be used to extract local features in pressure data, while RNN can be used to process the time series features of pressure data. Secondly, when calculating the historical matching value, a weighting mechanism can be introduced to assign weights to each dimension of the feature vector according to the importance of different features. For example, a higher weight can be given to feature dimensions that are highly correlated with leakage detection. In addition, a comprehensive evaluation can be combined with other features (such as the timestamp of the pressure data, the topological structure of the pipe network area, etc.) to improve the accuracy and reliability of the historical matching value. Finally, based on the generated historical matching value, the pressure-to-pressure matching adjustment amount can be further generated to provide a more accurate basis for leakage detection.

[0050] The above-mentioned embodiments of the present invention have the following beneficial effects: the technical solution can form a multi-dimensional analysis data set through intelligent segmentation processing based on the real-time collected pipeline pressure data, and establish a pressure feature model using the associated feature extraction technology. Through the phased pressure data retrieval and historical data dynamic update mechanism, the abnormal pressure fluctuations in the pipeline operation can be effectively identified, thereby improving the sensitivity of leakage detection. The pressure matching evaluation and aggregation processing method can eliminate data redundancy interference and ensure the reliability of the detection results.

[0051] This solution can accurately locate the leakage location with the help of pipeline area identification matching and feature vector embedding technology. By establishing a dynamic adjustment mechanism between historical matching values ​​and initial matching values, the leakage feature evaluation process can be optimized. By adopting the pressure correlation data generation and target pressure score calculation method, the leakage risk degree can be quantitatively evaluated, providing a scientific basis for pipeline maintenance decisions, and finally forming an orderly leakage feature sequence output to improve the intelligent level of water supply pipeline management.

[0052] like Figure 2 As shown, some embodiments of a water supply network leakage detection device include: The acquisition unit 201 is configured to acquire a pressure data set in response to receiving a pipeline network monitoring trigger signal; A segmentation processing unit 202 is configured to perform segmentation processing on the pressure data set to obtain a pressure data set and staged pressure data, wherein the staged pressure data includes a historical pressure data set; The correlation feature extraction unit 203 is configured to extract correlation features from each pressure data in the pressure data set to obtain a pressure feature data set; A retrieval processing unit 204 is configured to perform retrieval processing on the pressure feature data set based on the staged pressure data to obtain a target pressure data set; The updating processing unit 205 is configured to update the target pressure data set based on the historical pressure data set to obtain a target leakage feature sequence for output as a leakage detection result.

[0053] It is understandable that the modules described in the water supply network leakage detection device are similar to those described in the reference Figure 1 Therefore, the operations, features and beneficial effects described above for the water supply network leakage detection method are also applicable to the water supply network leakage detection device and the modules contained therein, and will not be repeated here.

[0054] Reference below Figure 3, which shows a schematic diagram of the structure of an electronic device 300 suitable for implementing some embodiments of the present invention. The electronic devices in some embodiments of the present invention may include but are not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 3 The terminal device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0055] like Figure 3 As shown, the electronic device 300 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 to a random access memory (RAM) 303. In the RAM 303, various programs and data required for the operation of the electronic device 300 are also stored. The processing device 301, the ROM 302, and the RAM 303 are connected to each other via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0056] Typically, the following devices may be connected to the I / O interface 305: an input device 306 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 308 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 309. The communication device 309 may allow the electronic device 300 to communicate with other devices wirelessly or by wire to exchange data. Figure 3 The electronic device 300 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead. Figure 3 Each block shown in the figure may represent one device, or may represent multiple devices as required.

[0057] Furthermore, the storage medium of the embodiment of the present application stores program instructions that can implement all the above methods, wherein the program instructions can be stored in the above storage medium in the form of a software product, including several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or terminal devices such as a computer, a server, a mobile phone, and a tablet.

[0058] The above descriptions are only some preferred embodiments of the present invention and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present invention is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above inventive concept. For example, the above features are replaced with (but not limited to) technical features with similar functions disclosed in the embodiments of the present invention.

Claims

1. A method for detecting leakage in a water supply network, comprising: In response to receiving a pipeline network monitoring trigger signal, acquiring a pressure data set; Segmenting the pressure data set to obtain a pressure data group set and staged pressure data, wherein the staged pressure data includes a historical pressure data set; Extracting associated features from each pressure data in the pressure data set to obtain a pressure feature data set; Based on the staged pressure data, the pressure characteristic data set is retrieved and processed to obtain a target pressure data set; Based on the historical pressure data set, the target pressure data set is updated to obtain a target leakage feature sequence for output as a leakage detection result.

2. The method according to claim 1, wherein: The extracting of associated features from each pressure data in the pressure data set to obtain a pressure feature data set includes: performing grouping processing on each pressure data in the pressure data group set to obtain an associated pressure data group set; performing redundancy filtering on each associated pressure data group in the associated pressure data group set to obtain a target associated pressure data group set; Performing matching evaluation on each target-associated pressure data group in the target-associated pressure data group set to obtain a pressure-to-pressure matching data set; Aggregation processing is performed on the pressure matching degree dataset to obtain a pressure feature dataset.

3. The method according to claim 2, wherein: The performing matching evaluation on each target-associated pressure data set in the target-associated pressure data set set to obtain a pressure matching data set includes: For each target-associated pressure data group in the target-associated pressure data group set, performing the following steps: generating an inter-pressure association data set based on the pressure data group set and the target-associated pressure data group; Based on the pressure-related data set, a matching evaluation is performed on the target-related pressure data set to obtain pressure-related matching data.

4. The method according to claim 3, wherein: The step of generating a pressure correlation data set based on the pressure data set set and the target-related pressure data set includes: Determine a piece of target-associated pressure data in the target-associated pressure data group that satisfies a first preset index condition as first pressure data; Determine a piece of target-associated pressure data in the target-associated pressure data group that satisfies a second preset index condition as second pressure data; For each pressure data group in the pressure data group set, the following steps are performed: determining a pressure data in the pressure data group that matches the first pressure data as target first pressure data; determining a piece of pressure data in the pressure data group that matches the second pressure data as target second pressure data; A matching degree evaluation is performed on the target first pressure data and the target second pressure data to obtain pressure correlation data.

5. The method according to claim 1, wherein: The updating process of the target pressure data set based on the historical pressure data set to obtain a target leakage feature sequence includes: For each target pressure data in the target pressure data set, the following steps are performed: extracting correlation features between the historical pressure data set and the target pressure data to obtain a set of pressure matching degree adjustment values; Determining the sum of the respective pressure matching degree adjustment amounts in the pressure matching degree adjustment amount set as the target pressure score; Determine the leakage characteristic data corresponding to the target pressure data and the target pressure score as target leakage characteristic data; The determined target leakage characteristic data are sorted to obtain a target leakage characteristic sequence.

6. The method according to claim 5, wherein: Each pressure data in the pressure data set includes a pipe network area identification group; And the extracting of associated features from the historical pressure data group and the target pressure data to obtain a pressure matching adjustment amount set includes: For each historical pressure data in the historical pressure data set, the following steps are performed: determining one pressure data in the pressure data set that matches the target pressure data as user pressure data; Performing a matching evaluation on the user pressure data and the historical pressure data to obtain a historical matching value; Performing a search process on the pressure feature data set to obtain a target initial matching value; Determining the historical pressure data, the historical matching value, and the target initial matching value as matching adjustment data; Based on the matching degree adjustment data, an adjustment amount of matching degree between pressures is generated.

7. The method according to claim 6, wherein: The performing matching evaluation on the user pressure data and the historical pressure data to obtain a historical matching value includes: From the pipe network area identification group included in the user pressure data, a pipe network area identification that matches the pipe network area identification group included in the historical pressure data is selected as a public area identification to obtain a public area identification group; Performing feature embedding processing on the user pressure data and the historical pressure data to obtain a target pressure feature vector and a historical pressure feature vector; A historical matching value is generated based on the common area identification group, the target pressure feature vector and the historical pressure feature vector.

8. A water supply network leakage detection device, comprising: an acquisition unit, configured to acquire a pressure data set in response to receiving a pipeline network monitoring trigger signal; a segmentation processing unit configured to perform segmentation processing on the pressure data set to obtain a pressure data group set and staged pressure data, wherein the staged pressure data includes a historical pressure data group; A correlation feature extraction unit is configured to extract correlation features from each pressure data in the pressure data set to obtain a pressure feature data set; A retrieval processing unit is configured to perform retrieval processing on the pressure characteristic data set based on the staged pressure data to obtain a target pressure data set; The update processing unit is configured to update the target pressure data set based on the historical pressure data set to obtain a target leakage feature sequence for output as a leakage detection result.

9. An electronic device, comprising: one or more processors; a storage device having one or more programs stored thereon, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 7.

10. A computer readable medium having a computer program stored thereon, wherein: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images