Water supply network leakage point positioning method and system, storage medium and computer

Abstract

The invention provides a water supply pipe network leakage point positioning method and system and a storage medium, and the positioning method comprises the steps: obtaining a topological structure of a water supply pipe network in a certain region, building a pipe network hydraulic model of the water supply pipe network according to the topological structure, and simplifying and checking a pipeline of the pipe network hydraulic model; according to the simplified and checked pipe network hydraulic model, carrying out regional division on the water supply pipe network, carrying out sensitive analysis on water pressure change of water supply nodes in the region, and determining detection points of each region of the water supply pipe network according to a sensitive analysis result; constructing a leakage point detection model based on the multilayer neural network, and performing model training on the leakage point detection model according to the flow and the water pressure of the water supply node; the water pressure data of the detection points corresponding to the different areas of the water supply network are obtained, the water pressure data of the detection points are imported into the leakage point detection model to be analyzed, the position of the leakage point in the water supply network is determined, and the water supply network leakage point positioning method is accurate in positioning and wide in application range.

Description

Technical Field

[0001] The present invention relates to the field of big data technology, and specifically to a method, system, storage medium and computer for locating leakage points in a water supply network. Background Art

[0002] As a water transportation carrier, water supply pipeline network is an important pillar of urban development. In recent years, water supply pipeline networks in many cities have experienced leakage due to long laying time, environmental corrosion, and imperfect information management of water systems. Since most water supply pipelines are buried underground and not noticed, a large amount of treated water is lost and wasted; this not only causes a waste of water resources, but also pollutes the water bodies in the pipeline network, and also causes hazards such as ground subsidence.

[0003] Currently, the commonly used methods for detecting pipeline leakage points mainly include passive leak detection and active leak detection. The passive leak detection method is to wait until the water leakage in the underground pipeline emerges from the ground and is discovered before leak detection is carried out. It requires less manpower and material resources but is also inefficient. The active leak detection method is mainly acoustic leak detection, which uses electronic equipment such as leak detectors. Leak detection personnel listen to the sound signals generated by pipeline leakage to check the leakage location in the pipeline. Since this method uses instrument listening, the environmental noise is large. The detection effect is closely related to the professional and technical level and engineering experience of the inspection personnel, and is highly subjective, making it difficult to promote on a large scale. Summary of the Invention

[0004] Based on this, the purpose of the present invention is to provide a method, system and storage medium for locating leakage points in a water supply network to solve the technical problems existing in the prior art.

[0005] The present invention provides a method for locating leakage points in a water supply network, comprising: Obtaining the topological structure of a water supply network in a certain area, building a hydraulic model of the water supply network based on the topological structure, and simplifying and verifying the pipes of the hydraulic model; Divide the water supply network into regions based on the simplified and verified hydraulic model of the network, perform sensitivity analysis on the water pressure changes of the water supply nodes in the region, and determine the detection points in each region of the water supply network based on the sensitivity analysis results; Building a leakage point detection model based on a multi-layer neural network, and training the leakage point detection model according to the flow rate and water pressure of the water supply node; The water pressure data of the detection points corresponding to different areas of the water supply network are obtained, and the water pressure data of the detection points are imported into the leakage point detection model for analysis to determine the location of the leakage point in the water supply network.

[0006] Preferably, the steps of obtaining the topological structure of the water supply network in a certain area, building a hydraulic model of the water supply network according to the topological structure, and simplifying and verifying the pipes of the hydraulic model include: Acquire attribute information and hydraulic information of the laid water supply network based on the monitoring system, wherein the attribute information and hydraulic information at least include pipes, valves, water pumps, and water sources; Importing the attribute information and hydraulic information into a hydraulic modeling platform, constructing a corresponding pipe network hydraulic model, obtaining the pipe structure characteristics of the pipe network hydraulic model, and classifying the complexity, density, and attributes of the pipes according to the pipe structure characteristics; Pipes with a complexity greater than a threshold are decomposed based on the flow direction of the pipeline connection lines and the pipeline water supply nodes, and pipelines with a density greater than a threshold and the same attributes are merged; The water distribution of the water supply node is adjusted according to the timeliness of the updated data of the monitoring system and the roughness of the inner wall of the pipeline, so as to perform model verification on the decomposed pipeline and the merged pipeline.

[0007] Preferably, the steps of dividing the water supply network into regions according to the simplified and verified hydraulic model of the network, performing sensitivity analysis on the water pressure changes of the water supply nodes in the regions, and determining the detection points in each region of the water supply network according to the sensitivity analysis results include: Inputting a certain initial flow rate into the water supply network, performing network adjustment processing on the water supply network according to the aggregation principle, and obtaining the water pressure value of each water supply node when the water supply network reaches initial equilibrium under the initial flow rate; The water supply network is divided into regions according to the water pressure value of each water supply node, wherein a water supply node belongs to a certain region, and the same region contains several water supply nodes whose water pressure difference meets the preset requirements; Obtaining the continuity equation and energy equation of each water supply node in a certain area, substituting the energy equation into the continuity equation, and obtaining the water pressure change sensitivity equation corresponding to each water supply node; Selecting a water supply node in a certain area as a variable water supply node, increasing a preset flow rate at the variable water supply node, and calculating the water pressure change of other water supply nodes except the variable water supply node according to the water pressure change sensitivity equation; The absolute values ​​of the water pressure changes of each water supply node are added together to determine the sensitivity of each water supply node in the selected area, the sensitivity of each water supply node is sorted, and the detection points in the selected area are determined according to the sorting results; Sensitivity analysis is performed on all areas divided by the water supply network in turn to determine all detection points in the water supply network.

[0008] Preferably, the step of constructing a leakage point detection model based on a multi-layer neural network and training the leakage point detection model according to the flow rate and water pressure of the water supply node includes: Constructing a leakage point detection model based on a back propagation neural network, and defining the leakage point detection model as a single output model; Constructing a leakage point test sample, wherein the leakage point test sample includes a plurality of test pipe sections having different leakage points, and an equal number of water pressure test points and water flow test points are arranged in each of the test pipe sections; Acquire test data of water pressure test points and water flow test points corresponding to all test pipe sections based on the same flow rate, and perform normalization processing on the test data; The normalized data is used as the input of the leakage point detection model, and the hidden layer water supply nodes of the back propagation neural network are determined according to the input layer water supply nodes and the output layer water supply nodes; Defining a transfer function, a learning rate, and a positioning error of the back propagation neural network, and performing iterative positioning calculation on the leakage point detection model according to the hidden layer water supply node, the transfer function, the learning rate, and the positioning error; When the error between the leakage point position predicted by iterative positioning and the actual leakage point position in the test pipe section is less than the positioning error, the iteration is stopped and the model training of the leakage point detection model is completed.

[0009] Preferably, the expression for the normalization process is:

[0010] Where, For the Normalized test data, For the The test data before normalization, 、 are the minimum and maximum values ​​of the test data respectively; The expression of the hidden layer water supply node is:

[0011] Where, is the number of water supply nodes in the hidden layer, is the number of water supply nodes in the input layer, is the number of water supply nodes in the output layer, is the adjustment factor; The iterative expression of the leakage point detection model is:

[0012] Where, 、 Leakage point detection model Hedi The predicted leakage point position at the iteration, is the Jacobian matrix of the backpropagation neural network optimization algorithm, is the damping factor, is the base of natural logarithms.

[0013] Preferably, the step of obtaining water pressure data of detection points corresponding to different areas of the water supply network, importing the water pressure data of the detection points into the leakage point detection model for analysis, and determining the location of the leakage point in the water supply network includes: Based on the hydraulic model of the pipe network, the simulated pressure value of each detection point when no leakage occurs in the water supply pipe network is obtained. Obtain the measured pressure value of each detection point in the water supply network; Constructing an objective function for leak point positioning according to the simulated pressure value and the measured pressure value; Initializing a particle swarm in a particle swarm algorithm, updating positions and velocities of particles in the particle swarm according to parameters of the initialized particle swarm, and solving the objective function based on the updated positions and velocities of the particles; Update the individual extreme value of the particle and the global optimal value of the particle according to the solution results, and determine whether all particles in the particle swarm have completed the update; If so, stop updating and find the target solution in the target function; If not, continue to update the positions and velocities of the particles in the particle swarm until all particles in the particle swarm are updated.

[0014] Preferably, the objective function of the leakage point location is expressed as:

[0015] Where, is the error optimization function, is the number of the detection point with leakage to be solved, To adjust the detection point of the analog pressure value, is the number of operating conditions, is the number of detection points, For the Under the working condition The simulated pressure value of each detection point, For the Under the working condition The measured pressure value of each detection point; The expression for particle update is:

[0016] Where, 、 The particles are Hedi The speed of updates, is the particle inertia weight, 、 They are the individual learning factor and social learning factor of particles, 、 Are all random numbers between 0-1, For the Update the individual extreme value of the particle, is the global optimal value of the particle, The current position of the particle.

[0017] The present invention also provides a water supply network leakage point positioning system, comprising: A construction module is used to obtain the topological structure of the water supply network in a certain area, build a water supply network hydraulic model based on the topological structure, and simplify and verify the pipes of the water supply network hydraulic model; a partitioning module for partitioning the water supply network into regions based on the simplified and verified hydraulic model of the network, performing a sensitivity analysis on the water pressure changes of the water supply nodes in the region, and determining the detection points in each region of the water supply network based on the sensitivity analysis results; A training module is used to build a leakage point detection model based on a multi-layer neural network, and to train the leakage point detection model according to the flow rate and water pressure of the water supply node; The analysis module is used to obtain water pressure data of detection points corresponding to different areas of the water supply network, import the water pressure data of the detection points into the leakage point detection model for analysis, and determine the location of the leakage point in the water supply network.

[0018] The present invention also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the above-mentioned method for locating leakage points in a water supply network is implemented.

[0019] The present invention also proposes a computer, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the above-mentioned method for locating leakage points in a water supply network is implemented.

[0020] Compared with the prior art, the present invention has the following beneficial effects: the method for locating leakage points in a water supply network provided by the present application first obtains the topological structure of the water supply network in a certain area, builds a hydraulic model of the water supply network based on the topological structure, and simplifies and verifies the pipes of the hydraulic model of the network. Simplifying and verifying the model can improve the accuracy of the model in simulating the operation of the real network, divide the water supply network into regions, and perform sensitivity analysis on the water supply nodes in the divided regions, reasonably determine the detection points, reduce investment while ensuring the detection requirements, construct a leakage point detection model and perform model training, obtain water pressure data of the detection points and import the trained leakage point detection model for analysis, and then determine the location of the leakage point in the water supply network. Constructing a leakage point detection model based on a multi-layer neural network for deep learning positioning can, on the one hand, detect leakage points in the underground water supply network in a timely manner, and on the other hand, effectively reduce the positioning error caused by subjective positioning of patrol personnel, which is suitable for large-scale promotion.

[0021] Additional aspects and advantages of the present invention will be set forth in part in the description which follows and, in part, will be obvious from the description which follows, or may be learned by practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a flow chart of the method for locating leakage points in a water supply network according to the first embodiment of the present invention; Figure 2 This is a structural block diagram of a computer in Embodiment 4 of the present invention.

[0023] The following specific embodiments will further illustrate the present invention in conjunction with the above-mentioned drawings. DETAILED DESCRIPTION

[0024] To facilitate understanding of the present invention, the present invention will be described more fully below with reference to the accompanying drawings. The drawings illustrate several embodiments of the present invention. However, the present invention may be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a more thorough and comprehensive understanding of the present invention.

[0025] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one skilled in the art to which this invention pertains. The terms used in this specification of the present invention are for the purpose of describing specific embodiments only and are not intended to limit the present invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0026] Example 1 See also Figure 1, which shows a method for locating a leakage point in a water supply network in a first embodiment of the present invention. Specifically, the method for locating a leakage point in a water supply network includes steps S10 to S40: S10, obtaining a topological structure of a water supply network in a certain area, building a hydraulic model of the water supply network based on the topological structure, and simplifying and verifying the pipes of the hydraulic model; Optionally, the model of the urban water supply network mainly includes a hydraulic model and a water quality model. The hydraulic model is mainly a network model constructed based on the hydraulic model and combined with the parameters of the actual network. It can be used for network optimization, intelligent zoning, network leak detection and other aspects.

[0027] In this embodiment, the steps of obtaining a topological structure of a water supply network in a certain area, building a hydraulic model of the water supply network based on the topological structure, and simplifying and verifying the pipes of the hydraulic model include: Acquire attribute information and hydraulic information of the laid water supply network based on the monitoring system, wherein the attribute information and hydraulic information at least include pipes, valves, water pumps, and water sources; Importing the attribute information and hydraulic information into a hydraulic modeling platform, constructing a corresponding pipe network hydraulic model, obtaining the pipe structure characteristics of the pipe network hydraulic model, and classifying the complexity, density, and attributes of the pipes according to the pipe structure characteristics; Pipes with a complexity greater than a threshold are decomposed based on the flow direction of the pipeline connection lines and the pipeline water supply nodes, and pipelines with a density greater than a threshold and the same attributes are merged; The water distribution of the water supply node is adjusted according to the timeliness of the updated data of the monitoring system and the roughness of the inner wall of the pipeline, so as to perform model verification on the decomposed pipeline and the merged pipeline.

[0028] In the specific implementation, the hydraulic information and attribute information of the pipeline network are first obtained based on the intelligent monitoring system, and imported into the hydraulic modeling platform to build a pipeline network hydraulic model corresponding to the actual water supply pipeline network. Common modeling platforms include WaterGEMS platform, InfoWorks Ws platform, etc. After the pipeline network hydraulic model is established, in order to avoid errors in the pipeline network data, the model data needs to be simplified and verified accordingly; the simplification of the pipeline network mainly includes the decomposition of complex pipeline networks and the merging of pipeline networks of the same type. The decomposition of complex pipeline networks is to decompose the complex pipeline network model into several independent pipeline network segments for hydraulic calculation. The merging of pipeline networks of the same type is to merge pipelines with dense distribution and the same attributes to reduce the influence of external factors such as the on-site environment and pipeline material on leakage location; the verification of the pipeline network is mainly to enable the established model to truly simulate the operation of the actual pipeline network, reduce the data collection delay in the modeling process, and the influence of different pipeline inner wall roughness caused by different pipeline materials on the operating status, and improve the accuracy of the model.

[0029] S20, dividing the water supply network into regions based on the simplified and verified hydraulic model of the network, performing sensitivity analysis on water pressure changes at water supply nodes within the regions, and determining detection points in each region of the water supply network based on the sensitivity analysis results; Optionally, the steps of dividing the water supply network into regions according to the simplified and verified hydraulic model of the network, performing sensitivity analysis on water pressure changes at water supply nodes within the regions, and determining detection points in each region of the water supply network according to the sensitivity analysis results include: Inputting a certain initial flow rate into the water supply network, performing network adjustment processing on the water supply network according to the aggregation principle, and obtaining the water pressure value of each water supply node when the water supply network reaches initial equilibrium under the initial flow rate; The water supply network is divided into regions according to the water pressure value of each water supply node, wherein a water supply node belongs to a certain region, and the same region contains several water supply nodes whose water pressure difference meets the preset requirements; Obtaining the continuity equation and energy equation of each water supply node in a certain area, substituting the energy equation into the continuity equation, and obtaining the water pressure change sensitivity equation corresponding to each water supply node; Selecting a water supply node in a certain area as a variable water supply node, increasing a preset flow rate at the variable water supply node, and calculating the water pressure change of other water supply nodes except the variable water supply node according to the water pressure change sensitivity equation; The absolute values ​​of the water pressure changes of each water supply node are added together to determine the sensitivity of each water supply node in the selected area, the sensitivity of each water supply node is sorted, and the detection points in the selected area are determined according to the sorting results; Sensitivity analysis is performed on all areas divided by the water supply network in turn to determine all detection points in the water supply network.

[0030] In this implementation, the water pressure in the pipeline network can not only reflect the operating status of the pipeline network, but also serve as the basis for pressure simulation. Theoretically, in order to truly reflect the operation of the pipeline network, measurements should be performed at each water supply node. It is understandable that measurement at each water supply node requires a large investment in manpower and material resources. Therefore, it is reasonable to select some representative detection points. The selection of detection points should at least meet the requirements of uniform distribution, strong representativeness, and convenient scheduling, which can reduce investment and ensure positioning effect.

[0031] To set up detection points, it is necessary to first divide the pipe network into zones. There is a one-to-many relationship between the zones and the water supply nodes. The pressures of the water supply nodes in the same zone are close, and the pressure change of a node in the pipe network will cause the pressure change of other nodes. Therefore, the water supply nodes in the zone are subjected to sensitivity analysis and sensitivity sorting to find the water supply node with the highest sensitivity. This water supply node can be the representative point in the current zone, that is, the detection point in the current zone. The sensitivity equation of the water supply node can be obtained by substituting the energy equation into the continuity equation and then expanding it. Sensitivity analysis is performed on all zones in turn to determine all the detection points in the water supply network.

[0032] S30, constructing a leakage point detection model based on a multi-layer neural network, and training the leakage point detection model according to the flow rate and water pressure of the water supply node; Optionally, the step of constructing a leakage point detection model based on a multi-layer neural network and training the leakage point detection model according to the flow rate and water pressure of the water supply node includes: Constructing a leakage point detection model based on a back propagation neural network, and defining the leakage point detection model as a single output model; Constructing a leakage point test sample, wherein the leakage point test sample includes a plurality of test pipe sections having different leakage points, and an equal number of water pressure test points and water flow test points are arranged in each of the test pipe sections; Acquire test data of water pressure test points and water flow test points corresponding to all test pipe sections based on the same flow rate, and perform normalization processing on the test data; The normalized data is used as the input of the leakage point detection model, and the hidden layer water supply nodes of the back propagation neural network are determined according to the input layer water supply nodes and the output layer water supply nodes; Defining a transfer function, a learning rate, and a positioning error of the back propagation neural network, and performing iterative positioning calculation on the leakage point detection model according to the hidden layer water supply node, the transfer function, the learning rate, and the positioning error; When the error between the leakage point position predicted by iterative positioning and the actual leakage point position in the test pipe section is less than the positioning error, the iteration is stopped and the model training of the leakage point detection model is completed.

[0033] The expression of the normalization process is:

[0034] Where, For the Normalized test data, For the The test data before normalization, 、 are the minimum and maximum values ​​of the test data respectively; The expression of the hidden layer water supply node is:

[0035] Where, is the number of water supply nodes in the hidden layer, is the number of water supply nodes in the input layer, is the number of water supply nodes in the output layer, is the adjustment factor; The iterative expression of the leakage point detection model is:

[0036] Where, 、 Leakage point detection model Hedi The predicted leakage point position at the iteration, is the Jacobian matrix of the backpropagation neural network optimization algorithm, is the damping factor, is the base of natural logarithms.

[0037] Optionally, a leakage point detection model is constructed based on a back propagation neural network. In the leakage point detection model, the output layer only has one leakage point layer. The back propagation neural network can perform back propagation deep learning iterations based on the gap between the calculated output result and the expected output value, and perform weight and threshold corrections in the error gradient descent method until the preset number of iterations; construct test samples. In this embodiment, the test samples can be six, and a leakage point is provided at different positions on each of the six test pipe sections. Each test pipe section is provided with a water pressure test point and five water flow test points. The six test samples are parallel experimental groups; the back propagation neural network The transfer function can be a Sigmoid function, a linear function, a Purelin function, etc. The input data range of the transfer function generally needs to be controlled between 0 and 1, so the collected test data needs to be normalized; the transfer function affects the mapping relationship between input and output, and the learning rate can be used to correct the weights and thresholds between neurons. If the learning rate is too large, the accuracy is insufficient, which easily leads to error omissions. If the learning rate is too small, it will lead to slow convergence; the positioning error reflects the accuracy of positioning. In this embodiment, the Sigmoid function is used as the transfer function, the learning rate is 0.05, and the positioning error is 0.005.

[0038] S40, obtaining water pressure data of detection points corresponding to different areas of the water supply network, importing the water pressure data of the detection points into the leakage point detection model for analysis, and determining the location of the leakage point in the water supply network.

[0039] The steps of obtaining water pressure data of detection points corresponding to different areas of the water supply network, importing the water pressure data of the detection points into the leakage point detection model for analysis, and determining the location of the leakage point in the water supply network include: Based on the hydraulic model of the pipe network, the simulated pressure value of each detection point when no leakage occurs in the water supply pipe network is obtained. Obtain the measured pressure value of each detection point in the water supply network; Constructing an objective function for leak point positioning according to the simulated pressure value and the measured pressure value; Initializing a particle swarm in a particle swarm algorithm, updating positions and velocities of particles in the particle swarm according to parameters of the initialized particle swarm, and solving the objective function based on the updated positions and velocities of the particles; Update the individual extreme value of the particle and the global optimal value of the particle according to the solution results, and determine whether all particles in the particle swarm have completed the update; If so, stop updating and find the target solution in the target function; If not, continue to update the positions and velocities of the particles in the particle swarm until all particles in the particle swarm are updated.

[0040] The objective function of the leakage point location is expressed as follows:

[0041] Where, is the error optimization function, is the number of the detection point with leakage to be solved, To adjust the detection point of the analog pressure value, is the number of operating conditions, is the number of detection points, For the Under the working condition The simulated pressure value of each detection point, For the Under the working condition The measured pressure value of each detection point; The expression for particle update is:

[0042] Where, 、 The particles are Hedi The speed of updates, is the particle inertia weight, 、 They are the individual learning factor and social learning factor of particles, 、 Are all random numbers between 0-1, For the Update the individual extreme value of the particle, is the global optimal value of the particle, The current position of the particle.

[0043] Furthermore, in order to determine the position of particles, quantum behavior is introduced into the conventional particle swarm algorithm. In quantum space, the position of particles depends on the wave function. Particles move in a one-dimensional potential well centered on the attraction point. The particle position update expression is:

[0044] Where, 、 The particles are Hedi The average position after updates, For the particle The wave function at the time of position update, is the contraction-expansion coefficient of quantum behavior, For the particle Hedi Real-time location between updates.

[0045] Optionally, in this embodiment, the simulated pressure value of each detection point and the measured pressure value of each detection point are obtained when no leakage occurs in the water supply network; when leakage occurs at a detection point in the water supply network, there is an error in the pressure parameter between the measured pressure value and the simulated pressure value of each detection point, and the objective function of the error is:

[0046] It can be seen from the objective function that after adjusting the simulated pressure value of a certain detection point k, the pressure error between the measured pressure value and the simulated pressure value of all detection points under different working conditions is calculated; the simulated pressure values ​​of all detection points are adjusted in turn, and the pressure error between the measured pressure value and the simulated pressure value of each detection point under different working conditions is compared when the simulated pressure values ​​of different detection points are adjusted. It can be understood that when the simulated pressure value of a certain detection point k is adjusted so that the pressure error between the measured pressure value and the simulated pressure value of all detection points under different working conditions is minimized, the detection point k can be considered as the actual leakage node; further, in this embodiment, the objective function is solved based on the particle swarm algorithm, and each particle transmits information with each other, so that the entire particle swarm shares each other's position and velocity information during the spatial operation process, and finally finds the optimal solution in space.

[0047] In summary, the method for locating leakage points in a water supply network provided by this application first obtains the topological structure of the water supply network in a certain area, builds a network hydraulic model of the water supply network based on the topological structure, and simplifies and verifies the pipes of the network hydraulic model. Simplifying and verifying the model can improve the accuracy of the model in simulating the operation of the real network, divide the water supply network into regions, and perform sensitivity analysis on the water supply nodes in the divided regions, reasonably determine the detection points, reduce investment while ensuring the detection requirements, construct a leakage point detection model and train the model, obtain the water pressure data of the detection point and import the trained leakage point detection model for analysis, and then determine the location of the leakage point in the water supply network. Constructing a leakage point detection model based on a multi-layer neural network for deep learning positioning can, on the one hand, detect leakage points in the underground water supply network in a timely manner, and on the other hand, effectively reduce the positioning error caused by the subjective positioning of patrol personnel, which is suitable for large-scale promotion.

[0048] Furthermore, in some optional embodiments, after determining the detection points in each area of ​​the water supply network, when a leak occurs in the pipeline, the upstream and downstream water flows will quickly squeeze toward the leak point and form a new steady state. The leak point will generate a negative pressure signal due to the pressure drop and propagate to both ends. Pressure sensors are arranged at both ends of the leak point, and the location of the leak point can be determined by monitoring the time difference between the pressure sensors receiving the negative pressure signal. Schematically, assuming that the detection points are numbered in sequence according to the direction of water flow, two pressure sensors are arranged at the detection points numbered 1 and 50. If the two pressure sensors do not detect a negative pressure signal, then there is no leak at the detection points 1 and 50. If the two pressure sensors detect a negative pressure signal before and after, then there is a leak at the detection points 1 and 50. The calculation expression for the leak point is:

[0049] Where, is the distance from the leakage point to a pressure sensor, D is the distance between the two sensors, is the propagation speed of negative pressure signal, - The time difference between the two pressure sensors detecting the negative pressure signal; Specifically, the method for locating the leakage point of the water supply network through the negative pressure signal includes: Obtaining negative pressure signals received by two pressure sensors; Establish a constrained variational model, import the negative pressure signal into the constrained variational model, introduce the Lagrangian function for solution, and convert the constrained variational problem into an unconstrained variational problem; The unconstrained variational problem is decomposed based on the variational mode to decompose the negative pressure signal into several fixed-bandwidth IMF components and high-frequency IMF components, remove the high-frequency IMF components in the negative pressure signal, and eliminate the noise in the negative pressure signal; A cross-correlation analysis is performed on the two negative pressure signals after noise elimination. The time difference between the two pressure sensors detecting the negative pressure signals is calculated based on the maximum peak in the cross-correlation time domain diagram. The leakage point is located according to the calculated time difference and the calculation expression of the leakage point.

[0050] The expression of the negative pressure signal received by the two pressure sensors is:

[0051]

[0052] Where, 、 are the negative pressure signals after noise reduction received by the two pressure sensors, 、 are the attenuation of the negative pressure signal when it is transmitted to the two pressure sensors, It is a pure negative pressure signal emitted by the leakage point. 、 are the times when the two pressure sensors receive negative pressure signals; The expression of negative pressure signal cross-correlation analysis is:

[0053] when hour, When the maximum value is reached, the time difference between the two negative pressure signals can be obtained from the cross-correlation time domain diagram.

[0054] Furthermore, in some optional embodiments, after determining the detection points in each area of ​​the water supply network, a high-voltage pulse signal is applied between the two detection points. If there is a leakage point between the two detection points, a discharge phenomenon will occur when the signal is transmitted to the leakage point due to the damage of the pipeline. In particular, for insulated rubber pipes or plastic pipes, the discharge phenomenon is more obvious. After the signal is discharged at the leakage point, it will release acoustic and electrical signals and magnetic field signals into the environment. Therefore, a sensor is used to pick up the acoustic and electrical signals on the path of laying the water supply pipeline, and the leakage between the two detection points can be located. Due to the influence of factors such as the surrounding medium and the depth of the pipeline network, the collected acoustic and electrical signals are often mixed with a large amount of environmental noise. It is difficult for staff to accurately locate the leakage point through the acoustic and electrical signals mixed with noise signals. The positioning accuracy mainly depends on technical level and engineering experience. The intensity of the pulsed magnetic field released around the pipeline network is much higher than the electromagnetic interference in the environment, and the propagation speed of the magnetic field signal is much higher than the propagation speed of the acoustic and electrical signals. The magnetic field signal can effectively remove the noise mixed in the acoustic and electrical signals, making it convenient for end users to locate the leakage point.

[0055] Specifically, the method for locating the leakage point of the water supply network by using magnetic field signals includes: Acquire acoustic and electrical signals and magnetic field signals along the water flow direction between the two detection points; The moment of receiving the magnetic field signal is used as the starting moment of collecting the acoustic and electrical signals, the discharge time at the leakage point of the high-voltage pulse signal is determined according to statistical laws, and the end moment of collecting the acoustic and electrical signals is determined according to the discharge time; The collected acoustic and electrical signals are divided into frames for processing. Each frame signal is a stable signal and the frames partially overlap to ensure a smooth transition between frames. The short-time energy threshold is set to calculate the short-time energy of the collected acoustic and electrical signals. Remove short-time energy acoustic and electrical signals greater than the short-time energy threshold to reduce noise in the environment, and send the denoised acoustic and electrical signals to the end user for leakage point fault location.

[0056] The expression for short-time energy calculation is:

[0057] Where, Indicates the The short-time energy of the frame acoustic signal, is the frame length, For time, For the Frame acoustic and electrical signals;

[0058] Where, is the window function for frame processing, is the frame shift length.

[0059] Example 2 This embodiment provides a water supply network leakage point locating system, including: A construction module is used to obtain the topological structure of the water supply network in a certain area, build a water supply network hydraulic model based on the topological structure, and simplify and verify the pipes of the water supply network hydraulic model; a partitioning module for partitioning the water supply network into regions based on the simplified and verified hydraulic model of the network, performing a sensitivity analysis on the water pressure changes of the water supply nodes in the region, and determining the detection points in each region of the water supply network based on the sensitivity analysis results; A training module is used to build a leakage point detection model based on a multi-layer neural network, and to train the leakage point detection model according to the flow rate and water pressure of the water supply node; The analysis module is used to obtain water pressure data of detection points corresponding to different areas of the water supply network, import the water pressure data of the detection points into the leakage point detection model for analysis, and determine the location of the leakage point in the water supply network.

[0060] Preferably, the steps of obtaining the topological structure of the water supply network in a certain area, building a hydraulic model of the water supply network according to the topological structure, and simplifying and verifying the pipes of the hydraulic model include: Acquire attribute information and hydraulic information of the laid water supply network based on the monitoring system, wherein the attribute information and hydraulic information at least include pipes, valves, water pumps, and water sources; Importing the attribute information and hydraulic information into a hydraulic modeling platform, constructing a corresponding pipe network hydraulic model, obtaining the pipe structure characteristics of the pipe network hydraulic model, and classifying the complexity, density, and attributes of the pipes according to the pipe structure characteristics; Pipes with a complexity greater than a threshold are decomposed based on the flow direction of the pipeline connection lines and the pipeline water supply nodes, and pipelines with a density greater than a threshold and the same attributes are merged; The water distribution of the water supply node is adjusted according to the timeliness of the updated data of the monitoring system and the roughness of the inner wall of the pipeline, so as to perform model verification on the decomposed pipeline and the merged pipeline.

[0061] Preferably, the steps of dividing the water supply network into regions according to the simplified and verified hydraulic model of the network, performing sensitivity analysis on the water pressure changes of the water supply nodes in the regions, and determining the detection points in each region of the water supply network according to the sensitivity analysis results include: Inputting a certain initial flow rate into the water supply network, performing network adjustment processing on the water supply network according to the aggregation principle, and obtaining the water pressure value of each water supply node when the water supply network reaches initial equilibrium under the initial flow rate; The water supply network is divided into regions according to the water pressure value of each water supply node, wherein a water supply node belongs to a certain region, and the same region contains several water supply nodes whose water pressure difference meets the preset requirements; Obtaining the continuity equation and energy equation of each water supply node in a certain area, substituting the energy equation into the continuity equation, and obtaining the water pressure change sensitivity equation corresponding to each water supply node; Selecting a water supply node in a certain area as a variable water supply node, increasing a preset flow rate at the variable water supply node, and calculating the water pressure change of other water supply nodes except the variable water supply node according to the water pressure change sensitivity equation; The absolute values ​​of the water pressure changes of each water supply node are added together to determine the sensitivity of each water supply node in the selected area, the sensitivity of each water supply node is sorted, and the detection points in the selected area are determined according to the sorting results; Sensitivity analysis is performed on all areas divided by the water supply network in turn to determine all detection points in the water supply network.

[0062] Preferably, the step of constructing a leakage point detection model based on a multi-layer neural network and training the leakage point detection model according to the flow rate and water pressure of the water supply node includes: Constructing a leakage point detection model based on a back propagation neural network, and defining the leakage point detection model as a single output model; Constructing a leakage point test sample, wherein the leakage point test sample includes a plurality of test pipe sections having different leakage points, and an equal number of water pressure test points and water flow test points are arranged in each of the test pipe sections; Acquire test data of water pressure test points and water flow test points corresponding to all test pipe sections based on the same flow rate, and perform normalization processing on the test data; The normalized data is used as the input of the leakage point detection model, and the hidden layer water supply nodes of the back propagation neural network are determined according to the input layer water supply nodes and the output layer water supply nodes; Defining a transfer function, a learning rate, and a positioning error of the back propagation neural network, and performing iterative positioning calculation on the leakage point detection model according to the hidden layer water supply node, the transfer function, the learning rate, and the positioning error; When the error between the leakage point position predicted by iterative positioning and the actual leakage point position in the test pipe section is less than the positioning error, the iteration is stopped and the model training of the leakage point detection model is completed.

[0063] Preferably, the expression for the normalization process is:

[0064] Where, For the Normalized test data, For the The test data before normalization, 、 are the minimum and maximum values ​​of the test data respectively; The expression of the hidden layer water supply node is:

[0065] Where, is the number of water supply nodes in the hidden layer, is the number of water supply nodes in the input layer, is the number of water supply nodes in the output layer, is the adjustment factor; The iterative expression of the leakage point detection model is:

[0066] Where, 、 Leakage point detection model Hedi The predicted leakage point position at the iteration, is the Jacobian matrix of the backpropagation neural network optimization algorithm, is the damping factor, is the base of natural logarithms.

[0067] Preferably, the step of obtaining water pressure data of detection points corresponding to different areas of the water supply network, importing the water pressure data of the detection points into the leakage point detection model for analysis, and determining the location of the leakage point in the water supply network includes: Based on the hydraulic model of the pipe network, the simulated pressure value of each detection point when no leakage occurs in the water supply pipe network is obtained. Obtain the measured pressure value of each detection point in the water supply network; Constructing an objective function for leak point positioning according to the simulated pressure value and the measured pressure value; Initializing a particle swarm in a particle swarm algorithm, updating positions and velocities of particles in the particle swarm according to parameters of the initialized particle swarm, and solving the objective function based on the updated positions and velocities of the particles; Update the individual extreme value of the particle and the global optimal value of the particle according to the solution results, and determine whether all particles in the particle swarm have completed the update; If so, stop updating and find the target solution in the target function; If not, continue to update the positions and velocities of the particles in the particle swarm until all particles in the particle swarm are updated.

[0068] Preferably, the objective function of the leakage point location is expressed as:

[0069] Where, is the error optimization function, is the number of the detection point with leakage to be solved, To adjust the detection point of the analog pressure value, is the number of operating conditions, is the number of detection points, For the Under the working condition The simulated pressure value of each detection point, For the Under the working condition The measured pressure value of each detection point; The expression for particle update is:

[0070] Where, 、 The particles are Hedi The speed of updates, is the particle inertia weight, 、 They are the individual learning factor and social learning factor of particles, 、 Are all random numbers between 0-1, For the Update the individual extreme value of the particle, is the global optimal value of the particle, The current position of the particle.

[0071] Example 3 This embodiment provides a storage medium having a computer program stored thereon. When the program is executed by a processor, the above-mentioned method for locating leakage points in a water supply network is implemented.

[0072] Example 4 The present invention also provides a computer, see Figure 2 , shown is a computer in an embodiment of the present invention, including a memory 10, a processor 20, and a computer program 30 stored in the memory 10 and executable on the processor 20. When the processor 20 executes the computer program 30, the above-mentioned method for locating leakage points in a water supply network is implemented.

[0073] The memory 10 includes at least one type of storage medium, including flash memory, a hard disk, a multimedia card, a card-type memory (e.g., SD or DX memory), a magnetic memory, a magnetic disk, an optical disk, etc. In some embodiments, the memory 10 may be an internal storage unit of a computer, such as the computer's hard disk. In other embodiments, the memory 10 may also be an external storage device, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a flash memory card, etc. Furthermore, the memory 10 may include both an internal storage unit of the computer and an external storage device. The memory 10 can be used not only to store application software installed in the computer and various types of data, but also to temporarily store data that has been output or is about to be output.

[0074] Among them, in some embodiments, the processor 20 can be an electronic control unit (Electronic Control Unit, abbreviated as ECU, also known as a vehicle computer), a central processing unit (CPU), a controller, a microcontroller, a microprocessor or other data processing chip, used to run the program code stored in the memory 10 or process data, such as executing access restriction programs.

[0075] It should be pointed out that Figure 2 The structure shown does not constitute a limitation of the computer. In other embodiments, the computer may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0076] Those skilled in the art will appreciate that the logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device), or in conjunction with such instruction execution system, apparatus, or device. For purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by an instruction execution system, apparatus, or device, or in conjunction with such instruction execution system, apparatus, or device.

[0077] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting, or processing it in another suitable manner as necessary, and then storing it in a computer memory.

[0078] It should be understood that various components of the present invention may be implemented using hardware, software, firmware, or a combination thereof. In the aforementioned embodiments, multiple steps or methods may be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one or a combination of the following technologies known in the art may be used: a discrete logic circuit having logic gate circuits for implementing logic functions on data signals, an application-specific integrated circuit having suitable combinational logic gate circuits, a programmable gate array (PGA), a field-programmable gate array (FPGA), etc.

[0079] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0080] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A method for locating leakage points in a water supply network, characterized in that: include: Obtaining the topological structure of a water supply network in a certain area, building a network hydraulic model of the water supply network according to the topological structure, and simplifying and verifying the pipelines of the network hydraulic model; Divide the water supply network into regions according to the simplified and verified hydraulic model of the network, perform sensitivity analysis on the water pressure changes of the water supply nodes in the region, and determine the detection points in each region of the water supply network according to the results of the sensitivity analysis; Building a leakage point detection model based on a multi-layer neural network, and training the leakage point detection model according to the flow and water pressure of the water supply node; The water pressure data of the detection points corresponding to different areas of the water supply network are obtained, and the water pressure data of the detection points are imported into the leakage point detection model for analysis to determine the location of the leakage point in the water supply network.

2. The method for locating leakage points in a water supply network according to claim 1, characterized in that: The steps of obtaining the topological structure of a water supply network in a certain area, building a water supply network hydraulic model according to the topological structure, and simplifying and verifying the pipelines of the water supply network hydraulic model include: Acquire attribute information and hydraulic information of the laid water supply network based on the monitoring system, wherein the attribute information and hydraulic information at least include pipes, valves, water pumps, and water sources; Importing the attribute information and hydraulic information into a hydraulic modeling platform, constructing a corresponding pipe network hydraulic model, obtaining the pipe structure characteristics of the pipe network hydraulic model, and classifying the complexity, density and attributes of the pipes according to the pipe structure characteristics; According to the flow direction of the pipeline connection line and the pipeline water supply node, the pipelines with a complexity greater than the complexity threshold are decomposed, and the pipelines with a density greater than the density threshold and the same attributes are merged; The water distribution of the water supply node is adjusted according to the timeliness of the updated data of the monitoring system and the roughness of the inner wall of the pipeline, so as to perform model verification on the decomposed pipeline and the merged pipeline.

3. The method for locating leakage points in a water supply network according to claim 1, characterized in that: The steps of dividing the water supply network into regions according to the simplified and verified hydraulic model of the network, performing sensitivity analysis on the water pressure changes of the water supply nodes in the region, and determining the detection points in each region of the water supply network according to the sensitivity analysis results include: Input a certain initial flow rate into the water supply network, perform network adjustment processing on the water supply network according to the aggregation principle, and obtain the water pressure value of each water supply node when the water supply network reaches initial balance under the initial flow rate; The water supply network is divided into regions according to the water pressure value of each water supply node, wherein a certain water supply node belongs to a certain region, and the same region contains several water supply nodes whose water pressure difference meets the preset requirements; Obtaining the continuity equation and energy equation of each water supply node in a certain area, substituting the energy equation into the continuity equation, and obtaining the water pressure change sensitivity equation corresponding to each water supply node; A water supply node in a certain area is selected as a variable water supply node, a preset flow is added to the variable water supply node, and the water pressure change of other water supply nodes except the variable water supply node is calculated according to the water pressure change sensitivity equation; The absolute values ​​of the water pressure changes of each water supply node are added to determine the sensitivity of each water supply node in the selected area, the sensitivity of each water supply node is sorted, and the detection point in the selected area is determined according to the sorting result; Sensitivity analysis is performed on all areas divided into the water supply network in turn to determine all detection points of the water supply network.

4. The method for locating leakage points in a water supply network according to claim 1, characterized in that: The step of constructing a leakage point detection model based on a multi-layer neural network and training the leakage point detection model according to the flow and water pressure of a water supply node comprises: Constructing a leakage point detection model based on a back propagation neural network, and defining the leakage point detection model as a single output model; Constructing a leakage point test sample, wherein the leakage point test sample includes a plurality of test pipe sections with different leakage points, and an equal number of water pressure test points and water flow test points are arranged in each of the test pipe sections; Acquire test data of water pressure test points and water flow test points corresponding to all test pipe sections based on the same flow rate, and normalize the test data; The normalized data is used as the input of the leakage point detection model, and the hidden layer water supply nodes of the back propagation neural network are determined according to the input layer water supply nodes and the output layer water supply nodes; Defining the transfer function, learning rate, and positioning error of the back propagation neural network, and performing iterative positioning calculation on the leakage point detection model according to the hidden layer water supply node, the transfer function, the learning rate, and the positioning error; When the error between the leakage point position predicted by iterative positioning and the actual leakage point position in the test pipe section is less than the positioning error, the iteration is stopped and the model training of the leakage point detection model is completed.

5. The method for locating leakage points in a water supply network according to claim 4, characterized in that: The expression of the normalization process is: In the formula, For the The normalized test data, For the The test data before normalization, , are the minimum and maximum values ​​of the test data respectively; The expression of the hidden layer water supply node is: In the formula, is the number of water supply nodes in the hidden layer, is the number of water supply nodes in the input layer, is the number of water supply nodes in the output layer, is the adjustment factor; The expression of the leakage point detection model iteration is: In the formula, , They are respectively the leakage point detection model and The predicted leakage point position at the iteration, is the Jacobian matrix of the back-propagation neural network optimization algorithm, is the damping factor, is the base of natural logarithms.

6. The method for locating leakage points in a water supply network according to claim 1, characterized in that: The step of obtaining water pressure data of detection points corresponding to different areas of the water supply network, importing the water pressure data of the detection points into the leakage point detection model for analysis, and determining the location of the leakage point in the water supply network includes: Based on the pipe network hydraulic model, the simulated pressure value of each detection point when no leakage occurs in the water supply pipe network is obtained. Obtain the measured pressure value of each detection point in the water supply network; Constructing an objective function for leak point positioning according to the simulated pressure value and the measured pressure value; Initializing a particle swarm in a particle swarm algorithm, updating positions and velocities of particles in the particle swarm according to the parameters of the initialized particle swarm, and solving the objective function based on the updated positions and velocities of the particles; Update the individual extreme value of the particle and the global optimal value of the particle according to the solution result, and judge whether all particles in the particle swarm have completed the update; If yes, stop updating and find the target solution in the target function; If not, continue to update the positions and velocities of the particles in the particle swarm until all particles in the particle swarm are updated.

7. The method for locating leakage points in a water supply network according to claim 6, characterized in that: The expression of the objective function of the leakage point positioning is: In the formula, is the error optimization function, is the number of the detection point with leakage to be solved, To adjust the detection point of the simulated pressure value, is the number of operating conditions, is the number of detection points, For the Under the working condition The simulated pressure value of each detection point, For the Under the working condition The measured pressure value of each detection point; The expression for particle update is: In the formula, , The particles are and The speed of updates, is the particle inertia weight, , They are the individual learning factor and social learning factor of particles, , are all random numbers between 0 and 1. For the Update the individual extreme value of the particle, is the global optimal value of the particle, The current position of the particle.

8. A water supply network leakage point positioning system, characterized in that: include; A construction module is used to obtain the topological structure of a water supply network in a certain area, build a network hydraulic model of the water supply network according to the topological structure, and simplify and check the pipelines of the network hydraulic model; A division module is used to divide the water supply network into regions according to the simplified and verified hydraulic model of the network, perform sensitivity analysis on the water pressure changes of the water supply nodes in the region, and determine the detection points of each region of the water supply network according to the sensitivity analysis results; A training module, used for building a leakage point detection model based on a multi-layer neural network, and performing model training on the leakage point detection model according to the flow rate and water pressure of the water supply node; The analysis module is used to obtain water pressure data of detection points corresponding to different areas of the water supply network, import the water pressure data of the detection points into the leakage point detection model for analysis, and determine the location of the leakage point in the water supply network.

9. A storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method for locating leakage points in a water supply network as described in any one of claims 1 to 7 is implemented.

10. A computer comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method for locating leakage points in a water supply network as described in any one of claims 1 to 7 is implemented.

Images