Water supply network leakage detection system and method based on ultrasonic water meter

Abstract

The invention relates to the technical field of water supply network leakage detection, in particular to a water supply network leakage detection system and method based on an ultrasonic water meter. The system obtains meter reading data of the ultrasonic water meter through a data acquisition module, and the existing ultrasonic water meter in a water supply network is adopted as a data source for water supply network leakage detection; the dependence on special leakage detection equipment is reduced, and the data acquisition cost is reduced; meanwhile, the data construction module is used for constructing a meter reading data completion matrix based on a sparse matrix completion method, so that the data demand is reduced, and the data acquisition cost is further reduced; besides, the leakage position of the water supply network can be judged by constructing a sensitivity matrix through the leakage analysis module and performing residual error calculation on the pipeline, so that a basis is provided for processing of maintenance personnel, and the working efficiency of the working personnel is improved.

Description

Technical Field

[0001] This application relates to the technical field of water supply network leakage detection, and particularly to a water supply network leakage detection system and method based on ultrasonic water meters. Background Art

[0002] Network leakage detection is a process of using specialized technical means to locate and evaluate liquid or gas leakage problems in water supply, gas supply and other pipe network systems caused by reasons such as pipeline aging, corrosion, damage or improper construction. This detection work is of great significance for ensuring the safe operation of the pipe network system, reducing resource waste and lowering operating costs. If the network leakage problem fails to be discovered and handled in time, it will not only cause serious waste of water resources or energy, increase the operating costs of water supply or gas supply enterprises, but also may trigger safety hazards such as ground subsidence and water quality pollution, posing a threat to public safety.

[0003] Traditional water supply network leakage detection and identification technologies mainly rely on physical detection equipment, such as listening rods, noise recorders, correlation leak detectors, etc., to identify leakage points through manual listening or instrument analysis. However, these traditional methods have high requirements for detection environmental conditions (such as noise level, pipeline burial depth), the quantity and layout of equipment, and the experience and skills of operators, which limit their detection efficiency and accuracy.

[0004] In view of this problem, the existing patent application with the publication number CN108984873A provides a real-time water supply network leakage detection method. This method realizes the comprehensive judgment of flow and pressure data by constructing a simulation model of the water supply network to be managed, combining the real-time monitoring data of pressure and flow, and comparing and analyzing the simulation results of the simulation model with the measured data of the actual network, so as to quickly locate the leakage position. This method significantly shortens the leakage detection cycle and improves the detection efficiency. However, this method of establishing a simulation model highly depends on the quality and quantity of the collected data, but obtaining high-quality monitoring data often requires a large amount of human, material and time costs, resulting in a high cost of obtaining data. Summary of the Invention

[0005] Based on this, it is necessary to provide a water supply network leakage detection system and method based on ultrasonic water meters for the problem of high data acquisition cost in the existing water supply network leakage detection methods.

[0006] In a first aspect, this application provides a water supply network leakage detection system based on ultrasonic water meters, and the system includes a data acquisition module, a data construction module and a leakage analysis module;

[0007] The data acquisition module is used to obtain the meter reading data of ultrasonic water meters and the topological structure data of the water supply network;

[0008] The data construction module is used to construct a meter reading data completion matrix based on the sparse matrix completion method, and construct a pipeline-node association matrix according to the topological structure data and a preset filling rule;

[0009] The leakage analysis module is used to calculate the sensitivity matrix of node flow according to the hydraulic model and the pipeline-node association matrix, calculate the residual of each pipeline according to the difference between the sensitivity matrix and the difference between the meter reading data completion matrix and the calculated value of the hydraulic model, and obtain the leakage detection result according to the residual of each pipeline.

[0010] Further, the data construction module uses the augmented Lagrangian function of the optimization problem for sparse matrix completion, and the expression of the augmented Lagrangian function is as follows:

[0011]

[0012] In the formula, X ∈ R m×n is the restored matrix, Z is the intermediate variable introduced to solve the optimization problem, with X = Z, T is the Lagrange multiplier matrix, ρ is a positive scalar, the symbol <> represents the inner product, and ‖.‖ F represents the F-norm, that is, the square root of the sum of the squares of all elements.

[0013] Further, the filling rule is as follows:

[0014]

[0015] In the formula, A is an n×m pipeline-node association matrix, n is the number of pipelines, and m is the number of nodes.

[0016] Further, the expression of the hydraulic model is:

[0017]

[0018] In the formula, B represents the hydraulic model, q is the pipeline flow, and h is the head loss.

[0019] Further, the sensitivity matrix includes the sensitivity matrix of node flow to node pressure and the sensitivity matrix of node flow to pipeline flow; among them, the construction of the sensitivity matrix of node flow to node pressure is as follows:

[0020]

[0021] In the formula, Q is the node flow, H is the node pressure, A is the pipeline-node association matrix, and B is the hydraulic model;

[0022] The construction of the sensitivity matrix of node flow to pipeline flow is as follows:

[0023]

[0024] In the formula, Q is the node flow, q is the pipe flow, A is the pipe-node incidence matrix, and B is the hydraulic model.

[0025] Further, the calculation formula of the residual is:

[0026]

[0027] In the formula, ΔH0 is the vector of pressure change monitored when leakage occurs, Δq0 is the vector of flow change monitored when leakage occurs; J H (Q l ) is the vector of pressure change monitored when leakage occurs in pipe l, J q (Q l ) is the vector of flow change monitored when leakage occurs in pipe l; W is the weight coefficient; is the residual of the pipe.

[0028] Further, the data acquisition module includes an ultrasonic water meter reading unit, a meter reading data collection unit, and a meter reading data parsing unit; the ultrasonic water meter reading unit is used to send a reading command to the ultrasonic water meter, the meter reading data collection unit is used to obtain the meter reading data sent by the ultrasonic water meter, and the meter reading data parsing unit is used to parse the meter reading data to obtain structured meter reading data.

[0029] Further, the system further includes a post-meter leakage volume calculation module; the post-meter leakage volume calculation module is used to calculate the mean and standard deviation of the minimum night flow value of the leakage pipe within a preset time period, and calculate the post-meter leakage value when both the mean and the standard deviation exceed the corresponding thresholds.

[0030] Further, the system further includes a leakage alarm module; the leakage alarm module is used to give a reminder when there is leakage in the water supply network and store the leakage data of the water supply network.

[0031] In a second aspect, the present application also provides a method for detecting leakage in a water supply network based on an ultrasonic water meter, and the method includes:

[0032] Step S1, obtaining the meter reading data of the ultrasonic water meter and the topological structure data of the water supply network;

[0033] Step S2, constructing a meter reading data completion matrix based on the sparse matrix completion method, and constructing a pipe-node incidence matrix according to the topological structure data and a preset filling rule;

[0034] Step S3, calculating the sensitivity matrix of the node flow according to the hydraulic model and the pipe-node incidence matrix;

[0035] Step S4, calculate the residual of each pipeline according to the difference between the sensitivity matrix and the difference between the meter reading data completion matrix and the calculated value of the hydraulic model;

[0036] Step S5, obtain the leakage detection result according to the residual of each pipeline.

[0037] The above water supply network leakage detection system and method based on ultrasonic water meters obtain the meter reading data of ultrasonic water meters through the data acquisition module, and use the existing ultrasonic water meters in the water supply network as the data source for water supply network leakage detection, reducing the dependence on special leakage detection equipment and the cost of obtaining data; at the same time, through the data construction module, the meter reading data is converted into a meter reading data completion matrix by using the sparse matrix completion method, reducing the data demand and further reducing the data acquisition cost; in addition, by constructing a sensitivity matrix and calculating the residual of the pipeline through the leakage analysis module, the leakage location of the water supply network can be judged, providing a basis for maintenance personnel to handle, and improving the work efficiency of the staff. Description of the Drawings

[0038] Figure 1 It is a schematic diagram of the application environment of the water supply network leakage detection system based on ultrasonic water meters in an embodiment;

[0039] Figure 2 It is a schematic diagram of the structure of the water supply network leakage detection system based on ultrasonic water meters in an embodiment;

[0040] Figure 3 It is a schematic diagram of the flow of the water supply network leakage detection method based on ultrasonic water meters in an embodiment. Detailed Embodiments

[0041] The water supply network leakage detection system based on ultrasonic water meters provided by this application realizes efficient connection and data interaction with ultrasonic water meters through the IOT (Internet of Things) platform. Specifically, as Figure 1 shown, an advanced NB-IOT (Narrowband Internet of Things) communication method is adopted between the ultrasonic water meter and the IOT platform. This communication method ensures the stability and reliability of data transmission with its characteristics of low power consumption, wide coverage, and large connection number. And between the IOT platform and the water supply network leakage detection system, an HTTP / MQ (Hypertext Transfer Protocol / Message Queue) communication method is adopted to further ensure the efficient and accurate transmission of data between different systems. In order to make the purpose, technical solution and advantages of this application clearer, the following further details this application in combination with the drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain this application and are not used to limit this application.

[0042] Embodiment 1

[0043] AsFigure 2 As shown in Figure 2 , the leakage detection system of the water supply pipe network based on the ultrasonic water meter in this embodiment includes a data acquisition module, a data construction module, and a leakage analysis module.

[0044] The data acquisition module is used to obtain the meter reading data of the ultrasonic water meter and the topological structure data of the water supply pipe network. Among them, the meter reading data includes flow data and pressure data. The topological structure data of the water supply pipe network is constructed by the water supply pipe network management and maintenance party according to the actual pipe network situation. In a preferred embodiment, the data acquisition module includes an ultrasonic water meter reading unit, a meter reading data collection unit, and a meter reading data analysis unit. The ultrasonic water meter reading unit is used to send a meter reading command to the ultrasonic water meter. The meter reading command is a standardized instruction including flow data acquisition parameters and pressure data acquisition parameters. Then, the meter reading command is sent to the IOT platform using the MQ or HTTP communication protocol. The IOT platform then sends the meter reading command to the corresponding ultrasonic water meter, and after the ultrasonic water meter obtains the meter reading command, it reports the meter reading data to the meter reading data collection unit. The meter reading data collection unit is used to obtain the meter reading data sent by the ultrasonic water meter and buffer and store the meter reading data. The meter reading data analysis unit is used to analyze the meter reading data to obtain structured meter reading data. In particular, the meter reading data can be a data frame, which is easy to analyze, thereby improving the data processing efficiency.

[0045] The data construction module is used to construct a meter reading data completion matrix based on the sparse matrix completion method, and construct a pipeline-node association matrix according to the topological structure data and the preset filling rules.

[0046] The meter reading data obtained from the ultrasonic water meter may contain a large number of zero values or missing values, and these data are organized in the form of a sparse matrix. A sparse matrix is a special matrix in which most elements are zero, which can effectively save storage space and improve data processing efficiency. For example, in a large water supply pipe network, there are a large number of water meters, but during a certain period, some water meters may have no data records. By constructing a sparse matrix, this data sparsity situation can be easily represented. Since there are a large number of missing values in the sparse matrix, in order to perform subsequent data analysis, it needs to be completed. The completion method can use interpolation algorithms, machine learning models, etc. For example, using time series analysis methods, according to the meter reading data of adjacent time points, the missing data is predicted and filled, so that the sparse matrix becomes a complete matrix for more accurate analysis. In this embodiment, the augmented Lagrangian function of the optimization problem is used for sparse matrix completion, and the expression of the augmented Lagrangian function is as follows:

[0047]

[0048] In the formula, X ∈ R m×nLet \(X\) be the recovery matrix, \(Z\) be the intermediate variable introduced to solve the optimization problem, with \(X = Z\), \(T\) be the Lagrange multiplier matrix, \(\rho\) be a positive scalar, the symbol \(\langle\cdot,\cdot\rangle\) denote the inner product, and \(\|\cdot\|\) F denote the F-norm, that is, the square root of the sum of the squares of all elements.

[0049] The meter reading data completion matrix includes a flow rate completion matrix and a pressure completion matrix. Specifically, after completing the sparse matrix completion for the flow rate data and pressure data, the flow rate completion matrix and the pressure completion matrix can be obtained. The flow rate completion matrix contains the flow rate data of each node or pipeline after completion, and the pressure completion matrix contains the corresponding pressure data.

[0050] The water supply network has a complex topological structure, including pipelines and nodes. Nodes are usually the locations where water meters are installed or the intersections of pipelines, and pipelines connect various nodes. The purpose of constructing the pipeline-node incidence matrix is to clarify the connection relationship between pipelines and nodes, so that the structure of the water supply network can be clearly understood, providing a basis for subsequent leakage analysis. For example, the rows of the matrix represent pipelines, the columns represent nodes, and the element values in the matrix represent the connection situation between pipelines and nodes. The element values can be determined according to a preset filling rule. For example, when the element value is 1, it means the pipeline is connected to the node, and 0 means the pipeline is not connected to the node. In this embodiment, the pipeline-node incidence matrix is constructed based on the filling rule shown in Equation (2),

[0051]

[0052] where \(A\) is an \(n\times m\) pipeline-node incidence matrix, \(n\) represents the number of pipelines, \(m\) represents the number of nodes, \(i = 1,2,\ldots,m\), \(j = 1,2,\ldots,n\).

[0053] The leakage analysis module is used to calculate the sensitivity matrix of node flow rates according to the hydraulic model and the pipeline-node incidence matrix, calculate the residual of each pipeline according to the difference between the sensitivity matrix and the difference between the meter reading data completion matrix and the calculated value of the hydraulic model, and obtain the leakage detection result based on the residual of each pipeline.

[0054] Among them, the expression of the hydraulic model is as shown in Equation (3):

[0055]

[0056] where \(B\) represents the hydraulic model, which is a diagonal matrix, \(q\) is the pipeline flow rate, \(h\) is the head loss, and 1.852 is the constant value in the Hazen-Williams formula.

[0057] The sensitivity matrix of node flow reflects the sensitivity of each node and pipeline in the water supply network to changes in flow and pressure. Among them, the sensitivity matrix of node flow includes the sensitivity matrix of node flow to node pressure and the sensitivity matrix of node flow to pipeline flow. The elements of the sensitivity matrix of node flow to node pressure are the change rates of node flow with respect to node pressure, and the elements of the sensitivity matrix of node flow to pipeline flow are the change rates of node flow with respect to pipeline flow. Specifically, the construction of the sensitivity matrix of node flow to node pressure is as follows:

[0058]

[0059] In the formula, Q is the node flow, H is the node pressure, A is the pipeline-node incidence matrix, and B is the hydraulic model.

[0060] The construction of the sensitivity matrix of node flow to pipeline flow is as follows:

[0061]

[0062] In the formula, Q is the node flow, q is the pipeline flow, A is the pipeline-node incidence matrix, and B is the hydraulic model. When there is a leakage in the water supply network, it will cause abnormal changes in flow and pressure. According to the sensitivity matrix, it can be analyzed which nodes and pipelines are the most sensitive to this abnormal change, so as to judge the possible leakage location.

[0063] Based on equations (4)-(5), the sensitivity matrix of node flow as shown in equation (6) is obtained:

[0064]

[0065] In the formula, i ∈ {1..n} represents the pipelines in the water supply network. Pipeline i connects two nodes a and b. represents the pressure change vector of node a when pipeline i leaks. represents the pressure change vector of node b when pipeline i leaks. represents the flow change vector of node a when pipeline i leaks. represents the flow change vector of node b when pipeline i leaks.

[0066] After that, calculate the difference between the meter reading data completion matrix and the calculated values of the hydraulic model to obtain the pressure change vector and the flow rate change vector monitored when leakage occurs. Among them, the meter reading data completion matrix contains the actual flow rate data and pressure data in the leakage state. The calculated values of the hydraulic model are the theoretical flow rate values and theoretical pressure values calculated or simulated by the hydraulic model under the preset node flow rate parameters. Exemplarily, when pipeline i connects two nodes a and b, first obtain the actual flow rate data and actual pressure data of nodes a and b from the meter reading data completion matrix, then input the preset node flow rate parameters into the hydraulic model for calculation or simulation to obtain the theoretical flow rate values and theoretical pressure values at nodes a and b under ideal conditions, that is, without leakage. Then, subtract the theoretical flow rate value of nodes a and b from the actual flow rate value to obtain the flow rate change vector, and subtract the theoretical pressure value of nodes a and b from the actual pressure value to obtain the pressure change vector.

[0067] Finally, calculate the residuals of each pipeline according to the formula shown in Equation (7). The residual of pipeline r specifically includes the pressure residual value ΔH r and the flow rate residual value Δq r :

[0068]

[0069] In the formula, ΔH0 is the pressure change vector monitored when leakage occurs, and Δq0 is the flow rate change vector monitored when leakage occurs; is the structured gradient vector, obtained according to the sensitivity matrix shown in Equation (6), where J H (Q l ) is the pressure change vector monitored when leakage occurs in pipeline l, and J q (Q l ) is the flow rate change vector monitored when leakage occurs in pipeline l; W is the weight coefficient; is the residual of the pipeline.

[0070] In this embodiment, the pipeline with the smallest residual is identified as the leakage pipeline, which can assist maintenance personnel to quickly find the leakage point for repair and improve work efficiency.

[0071] Furthermore, as Figure 1 shown, the water supply network leakage detection system based on ultrasonic water meters may further include a leakage alarm module. The leakage alarm module is used to give a reminder when there is leakage in the water supply network and store the water supply network leakage data. Among them, the water supply network leakage data includes the location of the pipeline where leakage occurs.

[0072] The water supply network leakage detection system based on ultrasonic water meters in this embodiment obtains the meter reading data of ultrasonic water meters through a data acquisition module. By using the existing ultrasonic water meters in the water supply network as the data source for water supply network leakage detection, it reduces the dependence on special leakage detection equipment and lowers the cost of obtaining data. At the same time, through the data construction module, the meter reading data is converted into a meter reading data completion matrix using the sparse matrix completion method, reducing the data demand and further lowering the data acquisition cost. In addition, through the leakage analysis module, a sensitivity matrix is constructed and the residual calculation of the pipeline is performed, which can determine the leakage location of the water supply network, thus providing a basis for maintenance personnel to handle and improving the work efficiency of the staff.

[0073] Embodiment 2

[0074] Based on Embodiment 1, as Figure 2 shown, the water supply network leakage detection system in this embodiment further includes a post-meter leakage volume calculation module; the post-meter leakage volume calculation module is used to calculate the mean and standard deviation of the minimum night flow values of the leakage pipelines within a preset time period, and calculate the post-meter leakage value when both the mean and the standard deviation exceed the corresponding thresholds.

[0075] Among them, the post-meter leakage volume is the leakage water volume that occurs in the pipeline part after the water meter. Specifically, by obtaining the flow data uploaded by the ultrasonic water meter connected to the leakage pipeline, the minimum night flow value of the leakage pipeline can be analyzed. Then, by continuously recording the minimum night flow value, a statistical histogram of the minimum night flow value is established, and the mean and standard deviation of the histogram are continuously updated. When both the mean and the standard deviation exceed the corresponding thresholds, the post-meter leakage value is calculated. Exemplarily, the post-meter leakage value can be the mean of the minimum night flow when there is leakage minus the mean of the minimum night flow under normal conditions.

[0076] Furthermore, the leakage alarm module in this embodiment is also used to summarize the post-meter leakage volume and send alarm information according to the severity of the leakage.

[0077] Specifically, after the leakage warning module obtains the post-meter leakage volume of the pipeline with leakage, it divides the pipelines with leakage into leakages of different severities according to the preset post-meter leakage volume thresholds of different levels, such as minor leakage, moderate leakage, and severe leakage. Further, different levels of warning information can be sent to maintenance personnel according to the severity of the leakage. Exemplarily, when minor leakage occurs, the leakage warning module sends a prompt message to the maintenance personnel to arrange regular inspections; when moderate leakage occurs, the leakage warning module sends a warning message, requiring the maintenance personnel to handle it as soon as possible; when severe leakage occurs, an emergency warning is sent to start an emergency handling process, which may include immediately dispatching personnel to the scene for emergency repair. The forms of sending warning information include but are not limited to multiple methods such as system pop-up windows, text messages, emails, and APP push to ensure that the warning information is conveyed in a timely manner. In addition, the leakage warning module can further classify leakages of different severities in combination with the duration of the leakage occurrence. For example, a long-term small amount of leakage can also be determined as severe leakage.

[0078] The water supply network leakage detection system based on ultrasonic water meters in this embodiment can monitor and analyze the mean and standard deviation of the minimum night flow of the leakage pipeline in real time through the post-meter leakage volume calculation module, and calculate the post-meter leakage value when both the mean and the standard deviation exceed the corresponding thresholds, which can provide a quantitative basis for leakage assessment and significantly improve the efficiency and accuracy of water supply network leakage detection; at the same time, by sending warning information according to the severity of the leakage through the leakage warning module, the work efficiency of maintenance personnel can be further improved.

[0079] Embodiment 3

[0080] Based on Embodiment 1 and Embodiment 2, as Figure 3 shown, this application also provides a method for detecting water supply network leakage based on ultrasonic water meters, including:

[0081] Step S1, obtaining the meter reading data of the ultrasonic water meter and the topological structure data of the water supply network.

[0082] Among them, the meter reading data includes high-precision and high-density flow data and pressure data.

[0083] Step S2, constructing a meter reading data completion matrix based on the sparse matrix completion method, and constructing a pipeline-node association matrix according to the topological structure data and the preset filling rules.

[0084] Specifically, first organize the meter reading data obtained in Step S1 into the form of a sparse matrix, and then use the augmented Lagrangian function shown in Equation (1) for sparse matrix completion to obtain the meter reading data completion matrix. The meter reading data completion matrix specifically includes a flow completion matrix and a pressure completion matrix. In addition, this embodiment constructs a pipeline-node association matrix using the filling rules shown in Equation (2).

[0085] Step S3: Calculate the sensitivity matrix of the node flow based on the hydraulic model and the pipeline-node incidence matrix.

[0086] Among them, the sensitivity matrix of the node flow is shown in Equation (6). The specific construction process can refer to Embodiment 1 and will not be elaborated here.

[0087] Step S4: Calculate the residual of each pipeline according to the difference between the sensitivity matrix, the meter reading data completion matrix and the calculated value of the hydraulic model. Among them, the calculation formula of the residual is shown in Equation (7).

[0088] Step S5: Obtain the leakage detection result according to the residual of each pipeline. In this embodiment, the pipeline with the smallest residual is regarded as the leaking pipeline.

[0089] Furthermore, the method for detecting water leakage in a water supply network based on ultrasonic water meters in this embodiment further includes:

[0090] Step S6: Calculate the mean and standard deviation of the minimum night flow value of the leaking pipeline within a preset time period, and calculate the post-meter leakage value when both the mean and the standard deviation exceed the corresponding thresholds.

[0091] Furthermore, the method for detecting water leakage in a water supply network based on ultrasonic water meters in this embodiment further includes:

[0092] Step S7: Give a reminder when there is water leakage in the water supply network and store the water leakage data of the water supply network.

[0093] The technical features of the above-described embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0094] The above-described embodiments only represent several implementation manners of the present application. Their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the invention patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several deformations and improvements can be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.

Claims

1. A water supply network leakage detection system based on ultrasonic water meter, characterized in that: The system includes a data acquisition module, a data construction module and a leakage analysis module; The data acquisition module is used to obtain meter reading data of the ultrasonic water meter and topological structure data of the water supply network; The data construction module is used to construct a meter reading data completion matrix based on a sparse matrix completion method, and to construct a pipeline-node association matrix according to the topological structure data and a preset filling rule; The leakage analysis module is used to calculate the sensitivity matrix of the node flow according to the hydraulic model and the pipeline-node association matrix, calculate the residual of each pipeline according to the difference between the sensitivity matrix and the meter reading data completion matrix and the hydraulic model calculation value, and obtain the leakage detection result according to the residual of each pipeline.

2. The water supply network leakage detection system based on ultrasonic water meter according to claim 1 is characterized in that: The data construction module uses the augmented Lagrangian function of the optimization problem to complete the sparse matrix. The expression of the augmented Lagrangian function is as follows: Where X∈R m×n is the recovery matrix, Z is the intermediate variable introduced to solve the optimization problem, X = Z, T is the Lagrange multiplier matrix, ρ is a positive scalar, the symbol <> represents the inner product, ||.|| F Represents the F-norm, which is the square root of the sum of all elements.

3. The water supply network leakage detection system based on ultrasonic water meter according to claim 1 is characterized in that: The filling rules are as follows: Where A is an n×m pipeline-node association matrix, n is the number of pipelines, and m is the number of nodes.

4. The water supply network leakage detection system based on ultrasonic water meter according to claim 1 is characterized in that: The expression of the hydraulic model is: Where B represents the hydraulic model, q is the pipe flow rate, and h is the head loss.

5. The water supply network leakage detection system based on ultrasonic water meter according to claim 1 is characterized in that: The sensitivity matrix includes a sensitivity matrix of node flow to node pressure and a sensitivity matrix of node flow to pipeline flow; wherein the sensitivity matrix of node flow to node pressure is constructed as follows: Wherein, Q is the node flow, H is the node pressure, A is the pipeline-node association matrix, and B is the hydraulic model; The sensitivity matrix of node flow to pipeline flow is constructed as follows: Wherein, Q is the node flow, q is the pipe flow, A is the pipe-node association matrix, and B is the hydraulic model.

6. The water supply network leakage detection system based on ultrasonic water meter according to claim 1 is characterized in that: The residual is calculated as follows: Where ΔH0 is the pressure change vector monitored when leakage occurs, and Δq0 is the flow change vector monitored when leakage occurs; J H (Q l ) is the pressure change vector monitored when leakage occurs in pipeline l, J q (Q l ) is the flow change vector monitored when leakage occurs in pipeline l; W is the weight coefficient; is the residual of the pipeline.

7. The water supply network leakage detection system based on ultrasonic water meter according to claim 1 is characterized in that: The data acquisition module includes an ultrasonic water meter reading unit, a meter reading data collection unit and a meter reading data parsing unit; the ultrasonic water meter reading unit is used to send a reading command to the ultrasonic water meter, the meter reading data collection unit is used to obtain the meter reading data sent by the ultrasonic water meter, and the meter reading data parsing unit is used to parse the meter reading data to obtain structured meter reading data.

8. The water supply network leakage detection system based on ultrasonic water meter according to claim 1 is characterized in that: The system also includes a post-meter leakage calculation module; the post-meter leakage calculation module is used to calculate the mean and standard deviation of the nighttime minimum flow value of the leaking pipeline within a preset time period, and calculate the post-meter leakage value when the mean and the standard deviation both exceed the corresponding threshold value.

9. The water supply network leakage detection system based on ultrasonic water meter according to claim 1 is characterized in that: The system also includes a leakage alarm module; the leakage alarm module is used to give a reminder when there is leakage in the water supply network and to store leakage data of the water supply network.

10. A water supply network leakage detection method based on ultrasonic water meter, characterized in that: The method comprises: Step S1, obtaining meter reading data of an ultrasonic water meter and topological structure data of a water supply network; Step S2, constructing a meter reading data completion matrix based on a sparse matrix completion method, and constructing a pipeline-node association matrix according to topological structure data and preset filling rules; Step S3, calculating the sensitivity matrix of the node flow according to the hydraulic model and the pipeline-node association matrix; Step S4, calculating the residual of each pipeline according to the sensitivity matrix and the difference between the meter reading data completion matrix and the hydraulic model calculation value; Step S5, obtaining leakage detection results according to the residual of each pipeline.

Images