A method, system, and device for detecting and locating water supply pipe leaks based on a probabilistic model.

Abstract

This invention provides a method, system, and device for detecting and locating leaks in water supply pipes based on a probabilistic model and nonlinear filtering. The method includes: acquiring information about the water supply pipeline, such as the length, diameter, friction, and elastic modulus of each pipe; constructing a pure hydraulic model based on the water supply network information; solving the continuity and momentum equations of the water supply network using the method of characteristics; establishing a probabilistic multiple leakage model based on the circumferential stress and yield stress of the pipes, combined with the solution of the differential equation system, to determine the location and quantity of leaks in the pipeline; and using nonlinear filtering to estimate the state space in a noisy environment to achieve real-time correction and prediction of leaks. This invention considers the situation of multiple leaks in the water supply network and the water pressure difference between horizontal and non-horizontal pipes. Starting from a pure hydraulic model, it effectively solves the original partial differential equation system using the method of characteristics, effectively reducing the high false alarm rate problem of deterministic models.

Description

Technical Field

[0001] This invention belongs to the field of municipal engineering and urban water supply network detection technology, specifically relating to a probabilistic model construction and leakage detection and location method, system and device for water supply pipes. Background Technology

[0002] Urban water supply networks are a crucial component of urban water supply systems and a core element of urban public utilities. With my country's urbanization, significant progress has been made in the construction of urban water supply networks, and the coverage rate of public water supply has continuously increased. However, because the early layout of water supply networks relied heavily on expert experience and lacked rigorous scientific basis, subsequent expansions and renovations easily lead to uneven water supply loads, resulting in leaks and even pipe bursts, causing serious economic losses and resource waste. Therefore, how to promptly detect and locate leaks, deploy manpower for maintenance and repair, minimize property and resource losses, and avoid safety hazards has become a critical issue.

[0003] With the rapid development of artificial intelligence technology, a series of leakage identification and location methods based on mathematical models and water supply pipe monitoring data have emerged. These methods have a wide detection range, are not easily affected by external factors, and can be used to assist traditional hardware-based methods. However, existing methods are usually based on deterministic hydraulic models, which may overestimate the possible location of pipeline leaks. Especially when there are non-horizontal pipes or multiple leaks in the pipeline network, the pipeline system will generate very noisy pressure signals over a large area. Existing models are difficult to process such signals, often resulting in false alarms or giving a large investigation range, which consumes a lot of manpower and resources. Summary of the Invention

[0004] To address the aforementioned technical problems in existing technologies, this invention provides a method, system, and device for detecting and locating water supply pipe leaks based on pipeline probabilistic modeling and nonlinear filtering. This invention considers the presence of multiple leaks in the water supply network and the water pressure difference between horizontal and non-horizontal pipes. Starting from a pure hydraulic model, it effectively solves the original partial differential equations using the method of characteristics. Combining the circumferential stress and yield stress of the pipeline, it provides a probabilistic leak detection model, effectively reducing the high false alarm rate problem of deterministic models and improving the accuracy and efficiency of pipeline network leak detection.

[0005] To achieve the above objectives, this invention provides a method for detecting and locating water supply pipe leaks based on a probabilistic model and nonlinear filtering, comprising: acquiring water supply pipe information, such as pipe length, pipe diameter, pipe friction, fluid elastic modulus, etc.; constructing a pure hydraulic model based on the water supply network information; solving the continuity equation and momentum equation of the water supply network using the method of characteristics; establishing a probabilistic multiple leakage model based on the circumferential stress and yield stress of the pipes, combined with the solution of the differential equation system, to determine the location and quantity of leaks in the pipeline; and using nonlinear filtering to estimate the state space under noisy conditions to achieve real-time correction and prediction of leaks. Specifically, the method includes the following steps:

[0006] A method for detecting and locating water supply pipe leaks based on a probabilistic model and nonlinear filtering, characterized by the following specific steps:

[0007] S1. Obtain information about the water supply network;

[0008] S2. Construct a hydraulic model based on the water supply network information;

[0009] S3. Solve the characteristic equations of the water supply network using the method of characteristics;

[0010] S4. Establish a probabilistic leakage model by solving the coupled differential equation system, including: establishing a probabilistic multiple leakage model based on the circumferential stress and yield stress of the pipeline and the solution of the differential equation system to determine the location and quantity of leakage in the pipeline.

[0011] S5. Nonlinear filtering is used to estimate the state space in a noisy environment, enabling real-time correction and prediction of leakage.

[0012] Furthermore, the water supply network information includes the pipe length, pipe diameter, pipe friction, and fluid elastic modulus of each pipe.

[0013] Furthermore, S2, based on the mathematical model of step S1, establish the characteristic equation, specifically including:

[0014] For any node n in the water supply network at time step k, the continuity equation and instantaneous variable equation in the hydraulic mathematical model are solved using the method of characteristics; among them, the characteristic equation corresponding to the continuity equation is shown in formula (1):

[0015]

[0016] in, These represent the outflow, inflow, and leakage flow at node n in the pipeline network at time k, respectively, in m³. 3 / s, the characteristic equation corresponding to the momentum equation is shown in formula (2):

[0017]

[0018] in, A is the cross-sectional area of ​​the pipe, in meters (m). 2 'a' is the wave velocity in the fluid medium inside the pipe, in m / s; 'g' is the acceleration due to gravity; λ n Set to a constant value of 0.001. and These are the coefficients of the positive and negative characteristic lines of the characteristic equation at time k, respectively; The pressure head at pipeline node n at time k is represented by the pressure head in meters (m).

[0019] Furthermore, S3, solve the characteristic equation of the water supply network using the method of characteristics, specifically including:

[0020] The characteristic coefficients along the positive and negative characteristic lines are shown in formulas (3) and (4), respectively:

[0021]

[0022]

[0023] in, D is the pipe diameter, in meters (m), and A is the pipe's cross-sectional area, in meters (m²). 2 f is the pipe friction coefficient, Δt is the time step, and |·| represents the absolute value of ·. Based on the pipe network conditions and the boundary conditions of the differential equations, the total energy head at node n can be obtained by solving the system of equations, as shown in formula (5).

[0024]

[0025] in, The coefficients representing the characteristic lines.

[0026] Furthermore, S4, the solution of the coupled differential equation system establishes a probabilistic multiple leakage model, specifically including: considering the water pressure changes caused by upstream and downstream pipeline scheduling, a probabilistic multiple leakage model is established, as shown in formula (6):

[0027]

[0028] Among them, P leak H represents the probability of a node n experiencing a loss. YS The yield stress coefficient of the pipeline; when the unsteady rise in water pressure caused by the scheduling generates circumferential or axial stress in the pipeline that exceeds 80% of the material's yield stress, the pipeline may leak or burst.

[0029] Furthermore, S5 estimates the state space under noisy conditions based on nonlinear filtering, and performs real-time correction and prediction of pipeline leakage, specifically including:

[0030] To monitor and predict the status of each node in the water supply network, assuming there are N nodes in the network, the state space description of the nonlinear Kalman filter method is shown in equation (7):

[0031]

[0032] in, These are the pressure heads at nodes 1 through N, respectively. These represent the inflow rates at nodes 1 through N, respectively. These represent the outflow rates at nodes 1 through N, respectively. These represent the leakage flow rates at nodes 1 through N; the model inputs are the upstream and downstream water head and the valve coefficient u. k =[H i H o C v ] T H i H represents the upstream head. o Representing downstream head, C v The valve coefficient is represented by the state-space form of the nonlinear stochastic difference equation describing the pipeline conditions, as shown in formula (8):

[0033] x k =f(x) k-1 ,u k-1 )+w k-1 (8)

[0034] Among them, w k-1 It is a preset Gaussian white noise, f(·) is the nonlinear equation describing the actual pipe flow change, (x k-1 ,u k-1 The system state and model input at time k-1 are used as inputs to the function f(·); they are linearized to obtain the Jacobian matrix J. x As shown in formula (9):

[0035]

[0036] Among them, for pipeline observation z k The observation equation is: Where v k It is preset Gaussian white noise; in addition, the initial conditions in the filter estimation are set based on the water supply pipe information.

[0037] A water supply pipe leakage detection and location system based on probabilistic models and nonlinear filtering, characterized in that it includes:

[0038] The pipeline monitoring data acquisition module is used to acquire pipeline-related data sent by the monitoring equipment. The monitoring data includes water level data, water flow data, water pressure data, pipeline circumferential stress, and yield stress of each node unit in the current monitoring area.

[0039] The pipeline leakage analysis result determination module is used to perform forecast evolution calculations and pipeline leakage analysis on monitoring point data to obtain the leakage analysis results for the current monitoring area.

[0040] The early warning module is used to generate early warning information and related data of leaking pipelines. The data includes pipeline data of the leak point, water flow data, water pressure data, and location data. The early warning information is then sent to the early warning device so that the early warning device can notify relevant maintenance personnel based on the early warning information.

[0041] A water supply pipe leakage detection and location device based on probabilistic model and nonlinear filtering, characterized in that it is configured according to the above-mentioned water supply pipe leakage detection and location system, including monitoring equipment, data acquisition equipment, data processing and analysis equipment, early warning equipment and terminal equipment;

[0042] The monitoring device, the data acquisition device, the early warning device, and the terminal device are all connected to the data processing and analysis device via signals.

[0043] The data processing and analysis equipment is used to execute the water supply pipe leakage detection and location method as described in claim 1, and to analyze the monitoring data through a probabilistic multiple leakage model.

[0044] Compared with the prior art, the beneficial effects of the present invention are reflected in:

[0045] 1. The detection and location method provided by this invention takes into account the multiple leakage situations in the water supply network and the water pressure difference between horizontal and non-horizontal pipes. Starting from a pure hydraulic model, it uses the method of characteristics to effectively solve the original partial differential equations. Combined with the circumferential stress and yield stress of the pipe, it provides a probabilistic leakage detection model, which effectively reduces the high false alarm rate problem of deterministic models.

[0046] 2. The hydraulic model establishment and solution steps in this invention have strong flexibility and can handle different application scenarios and pipeline distributions; the nonlinear filtering method solves the problem of network node state estimation under multiple leakage and high noise conditions, and can monitor pipeline conditions in real time, providing leakage detection and location services.

[0047] 3. This invention collects monitoring data through a pipeline monitoring module, automatically determines the pipeline leakage analysis results of the current monitoring pipeline network area and outputs early warning information through data processing and analysis equipment, and automatically notifies maintenance personnel through an alarm module, providing relevant data. This solves the problem that existing pipeline network monitoring methods rely heavily on manual labor, achieving integrated data collection, data analysis and early warning, assisting government departments in quickly and reasonably locating leakage points for repair, and reducing resource and economic losses.

[0048] 4. The data processing and analysis equipment of this invention analyzes monitoring data through a hydraulic probability model, which can realize intelligent data analysis, improve early warning accuracy and efficiency, and meet the needs of accurate and efficient monitoring of pipeline leakage. Attached Figure Description

[0049] Figure 1 This is a schematic flowchart of a water supply pipe leakage detection and location method provided in Embodiment 1 of the present invention;

[0050] Figure 2 This is a schematic diagram of the horizontal water supply pipeline in the residential area as described in Embodiment 1 of the present invention;

[0051] Figure 3 This is a schematic diagram of a water supply pipe leakage detection and location method provided in Embodiment 2 of the present invention;

[0052] Figure 4 This is a schematic diagram of a non-horizontal water supply pipeline in a mountainous area as described in Embodiment 2 of the present invention;

[0053] Figure 5(a) is a schematic diagram of the results of real-time monitoring of the pressure head of pipeline node 3 provided in Embodiment 2 of the present invention;

[0054] Figure 5(b) is a schematic diagram of the results of real-time monitoring of leakage flow at pipe node 3 provided in Embodiment 2 of the present invention;

[0055] Figure 6 This is a schematic diagram of the pipeline model described in this invention;

[0056] Figure 7 This is a schematic diagram of a water supply pipe leakage detection and location system provided in Embodiment 3 of the present invention;

[0057] Figure 8 This is a schematic diagram of a water supply pipe leakage detection and location device provided in Embodiment 4 of the present invention. Detailed Implementation

[0058] To make the technical solution of the present invention clearer, the technical solution provided by the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It is to be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention. Furthermore, for ease of description, only the parts related to the present invention are shown in the accompanying drawings, not all of the structures.

[0059] Example 1

[0060] refer to Figure 1 and Figure 2 This embodiment is applicable to the automatic processing of monitoring data from horizontal pipelines in urban residential areas, analyzing the monitoring data, and performing leak alarms and location. This method can be executed by a pipeline leak detection and location device, which can be implemented in software and hardware and is generally integrated into a data processing and analysis module or interrupt. See the attached diagram for a schematic of horizontal pipelines in urban residential areas. Figure 2 As shown.

[0061] A method for detecting and locating water supply pipe leaks based on probabilistic models and nonlinear filtering includes the following steps:

[0062] S110. Obtain water supply network information and establish a hydraulic mathematical model;

[0063] S111. In this embodiment, information such as pipe length, pipe diameter, pipe friction, and fluid elastic modulus of the pipeline network is first obtained. Then, the continuity equation and instantaneous variable equation of the water supply network are established based on the pipeline network information, as shown in the following formula (10):

[0064]

[0065] Where Q and H are the flow rate and head at a certain point in the pipeline, respectively, as dependent variables; t and x are time and spatial location, respectively, as independent variables. The purpose of solving the above system of equations is to determine the flow rate and head in the pipeline with respect to time and spatial location. Secondly, A is the cross-sectional area of ​​the pipeline, g is the acceleration due to gravity, f is the coefficient of friction of the pipeline, and a is the wave velocity in the fluid medium inside the pipeline, given by the following equation (11):

[0066]

[0067] Where c1 is a constant, typically taken as 1, assuming the pipe is always fixed by an expansion joint; e is the pipe wall thickness in meters (m); and β is the fluid elastic modulus of the pipe in Pascals (Pa), ρ = 1000 kg / m. 3 Where is the fluid density, E is the Young's modulus of the pipe, and D is the pipe diameter in meters.

[0068] S112. Solve the two hyperbolic governing equations for transient fluid flow and obtain accurate results using the method of characteristics. Based on the method of characteristics, the partial differential equations can be transformed into two ordinary differential equations:

[0069]

[0070]

[0071] S113, Numerical discretization of differential equation systems;

[0072] Integrating equation (12) along the positive characteristic line and solving for the flow at point P, we get:

[0073] Q P =C p -C a H(14)

[0074] in, Q A and H A Let be the known flow rate and pressure head at node A, and Δt be the integration time. Similarly, integrating equation (13) along the negative characteristic line and solving for the flow at point P yields:

[0075] Q P =C e +C a H(15)

[0076] in, Q B and H B The known flow rate and pressure head at node B are given, and the two unknowns are Q. P The values ​​of H can be determined by combining equations (14) and (15), that is,

[0077]

[0078] The value of H is similar. Therefore, by using formulas (14) and (15), the conditions for all interior points at the end of each time step can be determined.

[0079] S120, Probabilistic modeling of pipeline leakage points;

[0080] S121. Construct the boundary conditions for the internal points of pipeline leakage;

[0081] When a leak exists in a water supply pipeline, the transient fluid flow throughout the pipeline will be completely altered compared to a leak-free scenario. This example utilizes the orifice equation to simulate a leak at any internal location (node), where the continuity equation is implemented at the leak node. Figure 2 For example, for node 3, the pipeline model can be found here. Figure 6Assuming a leak is detected at node 3, the continuity equation and characteristic wave equation using the method of characteristics become:

[0082]

[0083]

[0084]

[0085] S122, Modeling of internal points of pipeline leakage;

[0086] For node 3 in the water supply network at time step k, the continuity equation is equation (17), and the characteristic equation is:

[0087]

[0088] in, λ3 = 0.001.

[0089] The characteristic coefficients along the positive and negative characteristic lines are as follows:

[0090]

[0091]

[0092] in, Δt is the time step. Therefore, the total energy head at node 3 can be obtained as follows:

[0093]

[0094] in,

[0095] S123, Probabilistic Modeling of Pipeline Leakage;

[0096] Considering the water pressure changes caused by upstream and downstream pipeline scheduling, a probabilistic model is established as follows:

[0097]

[0098] Among them, P leak H represents the probability of a leak occurring at this node. YS This refers to the yield stress coefficient of the pipeline. When the unsteady rise in water pressure caused by the scheduling results in circumferential or axial stress in the pipeline that exceeds 80% of the material's yield stress, the pipeline may experience leakage and rupture.

[0099] S130, Real-time prediction of water supply network leakage;

[0100] S131, State-space representation of the system;

[0101] by Figure 2 Taking a horizontal water supply network as an example, the state space description of the nonlinear Kalman filter method is as follows:

[0102]

[0103] The model inputs are upstream and downstream water head and valve coefficients:

[0104] u k =[H i H o C v ] T

[0105] The state-space form of the nonlinear stochastic difference equation describing the pipeline conditions is as follows:

[0106] x k =f(x) k-1 ,u k-1 )+w k-1 (8)

[0107] Among them, w k-1 The noise is Gaussian white noise. After first-order linearization, the required Jacobian matrix is:

[0108]

[0109] Secondly, the observation equation is expressed as

[0110]

[0111] Among them, v k It is Gaussian white noise.

[0112] S132, Initial covariance settings;

[0113] Before starting the estimation, the process error covariance Q needs to be set. k Measurement covariance R k The initial state x0 and the initial estimation error covariance P0 are also considered. The initial state estimate is determined through steady-state analysis, assuming zero leakage in the initial state of the water supply pipeline. The initial error covariance P0 is set to I. 24 C, where I 24 It is a 24×24 identity matrix, and C = 0.1 is a constant set based on the water supply network information. Secondly, the process error covariance Q... k for:

[0114]

[0115] Measurement covariance

[0116] S133, Model Prediction and Measurement Update;

[0117] The estimation process follows the time update and measurement update steps of Kalman filtering, as follows:

[0118]

[0119]

[0120] Among them, W k and V k These are the Jacobian matrices of the model nonlinear function and the measurement equation relative to the noise, respectively.

[0121] Example 2

[0122] Figure 3 This is a flowchart illustrating a method for detecting and locating leakage in non-horizontal pipelines in mountainous areas, as provided in Embodiment 2 of the present invention. This embodiment adds new steps to the above embodiments. Optionally, the method further includes: constructing boundary conditions for the upstream reservoir pipeline and the downstream valve. For parts not described in detail in this embodiment, please refer to Embodiment 1 above. See details. Figure 3 As shown in the diagram, see the schematic diagram of non-horizontal pipelines in mountainous areas. Figure 4 As shown, the method may include the following steps:

[0123] S210. Establishing a hydraulic mathematical model

[0124] S220, Probabilistic Modeling of Pipeline Leakage Internal Points

[0125] S221. Constructing boundary conditions for upstream reservoir pipelines

[0126] For the upstream (reservoir) of the pipeline (such as node 1), set boundary conditions.

[0127]

[0128] in, H R1 Let η be the pressure head of the reservoir, and η = 0.5 be the loss coefficient at the water inlet. With C 3,a Similarly.

[0129] S222, Constructing the boundary conditions for the downstream valve

[0130] The boundary conditions of the downstream hydraulic model are described using the valve position equations for steady-state flow through the valve. Taking node 7 as an example, the valve position differential equations for the downstream flow can be obtained by solving the following:

[0131]

[0132] in, H R2 For the downstream pressure head, Q s τ is the steady-state flow rate through the valve, and τ is the valve opening area. For a fully open valve, τ = 1.

[0133] S223, Boundary conditions for constructing internal points of pipeline leakage.

[0134] S224, Pipeline Leakage Internal Point Modeling

[0135] S225, Probabilistic Modeling of Pipeline Leakage

[0136] S230, Real-time prediction of water supply network leakage

[0137] Figure 6 This is a schematic diagram illustrating the detection and location results of a leak at pipe node 3, as provided in Example 2 of the present invention. A nonlinear filter is used to detect the flow rate and pressure head at the node in real time. Note that Example 2 is an extension of Example 1, and includes additional boundary conditions for upper and lower water levels.

[0138] Example 3

[0139] Figure 7 This is a schematic diagram of a water supply pipe leakage detection and location system provided in Embodiment 3 of the present invention. This system is configured within a data processing and analysis device. See also... Figure 7 The system includes: a pipeline monitoring data acquisition module, a pipeline leakage analysis result determination module, and an early warning module.

[0140] The pipeline monitoring data acquisition module is used to acquire monitoring data sent by the pipeline monitoring equipment. The monitored data includes water level data, water flow data, water pressure data, pipeline circumferential stress, and yield stress in the current monitoring area.

[0141] The pipeline leakage analysis result determination module is used to perform forecast evolution calculations and pipeline leakage analysis on monitoring point data based on hydraulic mathematical models, and obtain the leakage analysis results of the current monitoring area;

[0142] The early warning module is used to generate early warning information, including pipeline data, water flow data, water pressure data and location data of the leakage point, if it is determined that the pipeline leakage analysis result exceeds the early warning threshold set in the probability model, and send the alarm information to the early warning device so that the early warning device can notify relevant maintenance personnel based on the early warning information.

[0143] Based on the above technical solutions, the pipeline leakage analysis result determination module is further used to perform evolution calculations on the pipeline water flow based on the hydraulic probability model to obtain the water flow evolution results of each node unit watershed, wherein the water flow evolution results include at least one of the average flow velocity, flow rate, water pressure, and flow rate of each node unit watershed in the current monitoring area.

[0144] Based on the above technical solutions, the pipeline leakage analysis result determination module is also used to perform pipeline leakage analysis on the water flow evolution results based on the pipeline hydraulic model, and to calculate the pipeline leakage analysis results of the current monitoring area using the finite volume method. The pipeline leakage analysis results include at least the flow velocity, water pressure and location of each node unit watershed in the current monitoring area.

[0145] Based on the above technical solutions, the system further includes: a sending and receiving module; wherein, the sending module is used to send the monitoring data and the leakage analysis results to the terminal device; and the receiving module is used to receive the leakage analysis results from the pipeline leakage analysis result determination module.

[0146] The technical solution provided in this embodiment collects monitoring data through a pipeline monitoring module, automatically determines the pipeline leakage analysis results for the current monitored pipeline network area and outputs early warning information through data processing and analysis equipment, and automatically notifies maintenance personnel through an alarm module, providing relevant data. This solves the problem of existing pipeline network monitoring methods relying heavily on manual labor, achieving integrated data collection, analysis, and early warning. It assists government departments in quickly and rationally locating leakage points for repair, reducing resource and economic losses. The data processing and analysis equipment analyzes the monitoring data using a hydraulic probability model, enabling intelligent data analysis, improving early warning accuracy and efficiency, and meeting the need for precise and efficient pipeline leakage monitoring.

[0147] Example 4

[0148] Figure 8 This is a schematic diagram of a water supply pipe leakage detection and location device provided in Embodiment 4 of the present invention. (See also...) Figure 8 As shown, the device includes: pipeline monitoring equipment, data acquisition equipment, data processing and analysis equipment, early warning equipment, and terminal equipment.

[0149] The data processing and analysis device is used to acquire monitoring data sent by the monitoring device. The monitoring data includes water level data, water flow data, water pressure data, pipeline circumferential stress, and yield stress of each node unit in the current monitoring area.

[0150] Based on the hydraulic probability model and nonlinear filtering method, the monitoring data is used to perform forecast evolution calculation and leakage analysis to obtain the leakage analysis results of each node unit in the current monitoring area.

[0151] If the leakage analysis results are determined to exceed the set warning threshold, a warning message is generated, including leakage node pipe data, water flow data, water pressure data, and location data, and the warning message is sent to the warning device so that the warning device can issue a warning based on the warning message.

[0152] The pipeline monitoring equipment is used to collect detection data in non-digital signal format within the current monitoring node area and send the non-digital signal format detection data to the data acquisition equipment;

[0153] The data acquisition device is used to convert the non-digital signal format pipeline monitoring data into digital signal format pipeline monitoring data.

[0154] The early warning device is used to receive early warning information sent by the data processing and analysis device, and generate prompt information based on the early warning information;

[0155] The terminal device is used to receive the monitoring data and pipeline leakage analysis results sent by the data processing and analysis device, and to visualize the monitoring data and pipeline leakage analysis results.

[0156] It should be noted that in the above embodiments of the pipeline leakage detection and location system, the various modules are divided according to functional logic and are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy differentiation and are not used to limit the scope of protection of the present invention.

[0157] Finally, the above are merely preferred embodiments and technical principles of the present invention. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments, and may include many other equivalent embodiments without departing from the concept of the present invention, the scope of which is determined by the scope of the appended claims.

Claims

1. A method for detecting and locating water supply pipe leakage based on a probabilistic model and nonlinear filtering, characterized in that, The specific steps include the following: S1. Obtain information about the water supply network; S2. Construct a hydraulic model based on the water supply network information; S3. Solve the characteristic equations of the water supply network using the method of characteristics; S4. Establish a probabilistic leakage model by solving the coupled differential equation system, including: establishing a probabilistic multiple leakage model based on the circumferential stress and yield stress of the pipeline and the solution of the differential equation system to determine the location and quantity of leakage in the pipeline. S5. Using nonlinear filtering to estimate the state space in a noisy environment to achieve real-time correction and prediction of leakage; Step S5 specifically includes: To monitor and predict the status of each node in the water supply network, it is assumed that the water supply network has a total of The state-space description of the nonlinear Kalman filter method for each node is shown in Equation (7): （7） in, They are nodes The pressure head at the location, They are nodes Inflow at the location, They are nodes Outflow rate at the location They are nodes The leakage flow rate at the point; the model inputs are the upstream and downstream water head and the valve coefficient. ,in Representing the upstream water head, Represents the downstream head. The valve coefficient is represented by the state-space form of the nonlinear stochastic difference equation describing the pipeline condition, as shown in formula (8): （8） in, It is preset Gaussian white noise. This is a nonlinear equation describing the changes in water flow in a pipeline in practice. for The system state and model input at time t are used as functions. The input is linearized to obtain the Jacobian matrix. As shown in formula (9): （9） Among them, pipeline observation The observation equation is: ,in It is preset Gaussian white noise; in addition, the initial conditions in the filter estimation are set based on the water supply pipe information.

2. The method for detecting and locating water supply pipe leakage based on probabilistic models and nonlinear filtering as described in claim 1, characterized in that: The water supply network information includes the length, diameter, friction, and fluid elastic modulus of each pipe.

3. The method for detecting and locating water supply pipe leakage based on probabilistic models and nonlinear filtering as described in claim 1, characterized in that: S2. Establish the characteristic equation based on the mathematical model in step S1, specifically including: for Any node in the water supply network at a time step The continuity equation and instantaneous variable equation in the hydraulic mathematical model are solved using the method of characteristics; the characteristic equation corresponding to the continuity equation is shown in formula (1): （1） in, , , They are respectively Time-based pipeline nodes Outflow, inflow and leakage, in units of The characteristic equation corresponding to the momentum equation is shown in formula (2): （2） in, , It is the cross-sectional area of ​​the pipe, in units of... , It is the wave velocity within the fluid medium inside the pipe, measured in units of... , It is gravitational acceleration. Set to a constant value of 0.

001. and These are the characteristic equations. The coefficients of the positive and negative characteristic lines at time points; represent Time-based pipeline nodes Pressure head, unit is .

4. The method for detecting and locating water supply pipe leakage based on probabilistic models and nonlinear filtering as described in claim 3, characterized in that: S3. Solve the characteristic equations of the water supply network using the method of characteristics, specifically including: The characteristic coefficients along the positive and negative characteristic lines are shown in formulas (3) and (4), respectively: （3） （4） in, , It is the pipe diameter, in units of , It is the cross-sectional area of ​​the pipe, in units of... , It is the coefficient of friction of the pipeline. It is the time step. express The absolute value; represent Time-based pipeline nodes Outflow rate, in units of , represent -1 Time Network Node Pressure head, unit is ; Based on the pipeline network conditions and the boundary conditions of the differential equations, a system of simultaneous equations can be established to obtain the node... The total energy head at the point is shown in formula (5): （5） in, , representing the coefficient of the characteristic line.

5. The method for detecting and locating water supply pipe leakage based on probabilistic models and nonlinear filtering as described in claim 4, characterized in that: S4. The solution of the coupled differential equation system establishes a probabilistic multiple leakage model, specifically including: considering the water pressure changes caused by upstream and downstream pipeline scheduling, a probabilistic multiple leakage model is established, as shown in formula (6): （6） in, For nodes The probability of leakage. The yield stress coefficient of the pipeline; when the unsteady rise in water pressure caused by the scheduling generates circumferential or axial stress in the pipeline that exceeds 80% of the material's yield stress, the pipeline may leak or burst.

6. A water supply pipe leakage detection and location system based on a probabilistic model and nonlinear filtering, implementing the water supply pipe leakage detection and location method as described in claim 1, characterized in that, include: The pipeline monitoring data acquisition module is used to acquire pipeline-related data sent by the monitoring equipment. The monitoring data includes water level data, water flow data, water pressure data, pipeline circumferential stress, and yield stress of each node unit in the current monitoring area. The pipeline leakage analysis result determination module is used to perform forecast evolution calculations and pipeline leakage analysis on monitoring point data to obtain the leakage analysis results for the current monitoring area. The early warning module is used to generate early warning information and related data of leaking pipelines. The data includes pipeline data of the leak point, water flow data, water pressure data, and location data. The early warning information is sent to the early warning device so that the early warning device can notify relevant maintenance personnel based on the early warning information. The system includes a sending and receiving module; wherein the sending module is used to send the monitoring data and the leakage analysis results to the terminal device; and the receiving module is used to receive the leakage analysis results from the pipeline leakage analysis result determination module.

7. A water supply pipe leakage detection and location device based on probabilistic model and nonlinear filtering, characterized in that, The water supply pipe leakage detection and location method according to claim 2 is configured to include monitoring equipment, data acquisition equipment, data processing and analysis equipment, early warning equipment, and terminal equipment; The monitoring device, the data acquisition device, the early warning device, and the terminal device are all connected to the data processing and analysis device via signals. The data processing and analysis equipment is used to execute the water supply pipe leakage detection and location method as described in any one of claims 1 to 5, and to analyze the monitoring data through a probabilistic multiple leakage model.

Images