Method and system for monitoring leakage of urban water supply pipe network

Abstract

The invention provides an urban water supply pipe network leakage monitoring method, relates to the technical field of urban water supply pipe networks, and aims to reduce dependence on additional equipment and data and improve estimation precision. The method comprises the steps that a dynamic water supply pipe network model is constructed, and leakage flow and the water consumption for metering DMA in a partition mode under the leakage working condition are obtained through simulation based on the constructed dynamic water supply pipe network model; constructing a neural network water consumption prediction model, and predicting normal water consumption when no leakage occurs through training learning; a neural network leakage monitoring model is constructed, and the neural network leakage monitoring model takes normal water consumption and water consumption of DMA subarea metering under the leakage working condition as input parameters to judge whether leakage occurs in each DMA subarea or not; and for the DMA partition with the leakage, establishing a leakage positioning model by taking the minimum square error of the monitoring value and the simulation value of the pressure measuring point when the leakage occurs as an objective function, and solving the leakage positioning model to position the leakage position of the DMA partition.

Description

Technical Field

[0001] The present invention relates to the technical field of urban water supply pipe networks, and more specifically, to a method and system for monitoring leakage of urban water supply pipe networks. Background Art

[0002] With the acceleration of urbanization, urban water supply networks, as an important part of urban infrastructure, are receiving more and more attention for their safety and efficiency. However, the leakage problem of water supply networks has always been a major problem plaguing urban water management. Traditional leakage detection methods such as the independent metering area DMA night minimum flow method, hydraulic model method, equipment method, etc. have problems such as high cost, low efficiency, and reliance on a large number of additional equipment and data, which are difficult to meet the needs of efficient, intelligent, and precise management of modern urban water supply systems.

[0003] In recent years, the advantages of neural networks in complex system modeling and prediction have gradually emerged, providing new ideas and methods for water supply network leakage detection. By simulating the working mode of human brain neurons, neural networks can automatically learn the complex relationship between input and output and achieve efficient prediction and classification tasks.

[0004] Although the leakage detection method based on neural network has been initially applied, it still has the following shortcomings:

[0005] Insufficient accuracy in estimating legal water consumption: Traditional DMA nighttime minimum flow method and other methods rely on accurate estimation of legal water consumption. However, in reality, legal water consumption is affected by many factors and is difficult to estimate accurately, resulting in low accuracy in leakage detection.

[0006] Reliance on additional equipment and data: Some neural network leak detection methods require the installation of additional flow, pressure, acoustic and other equipment, which increases cost and complexity.

[0007] Insufficient model generalization ability: Some models are designed only for specific scenarios or data sets, and it is difficult to adapt to the leakage detection needs of different regions and different types of pipeline networks. Summary of the invention

[0008] The purpose of the present invention is to provide a method and system for monitoring leakage in urban water supply networks, so as to solve the problems in the prior art of insufficient accuracy in estimating legal water consumption, reliance on additional equipment and data, and insufficient model generalization ability.

[0009] In order to achieve the above object, the present invention provides a method for monitoring leakage of urban water supply pipe network, the method comprising:

[0010] Step S1: construct a dynamic water supply network model, and simulate and obtain the leakage flow and the water consumption of the DMA under leakage conditions based on the constructed dynamic water supply network model;

[0011] Step S2: constructing a neural network water consumption prediction model, wherein the neural network water consumption prediction model uses historical water consumption data, weather conditions and holiday information as input parameters, and predicts normal water consumption when no leakage occurs through training and learning;

[0012] Step S3: constructing a neural network leakage monitoring model, wherein the neural network leakage monitoring model uses the normal water consumption output in step S2 and the water consumption of the DMA under leakage conditions as input parameters to determine whether leakage occurs in each DMA partition;

[0013] Step S4: For the DMA partition where leakage occurs, a leakage location model is established with the minimum square error between the monitored value and the simulated value of the pressure measuring point when leakage occurs as the objective function, and the leakage location model is solved to locate the leakage position of the DMA partition.

[0014] Preferably, the step S1 comprises:

[0015] Step S1.1: Use EPANET software model to build a dynamic water supply network model;

[0016] Step S1.2: Setting up a diffuser on the EPANET software model node to simulate a leakage accident;

[0017] Step S1.3: calling the dynamic link library of the EPANET software model to calculate the leakage flow;

[0018] Step S1.4: In the dynamic water supply network model, metering water meters are set on the water supply pipes at the boundaries of each DMA partition to simulate the actual water metering process. According to the principle of water balance, the water volume difference between the inflow and outflow of each DMA partition is calculated to obtain the water consumption of the partition metering DMA under the leakage condition.

[0019] Preferably, the calculation formula of the leakage flow in step S1.3 satisfies:

[0020]

[0021] Among them, q is the leakage flow through the diffuser, μ is the flow coefficient, which is used to characterize the severity of pipeline leakage, p is the pressure at the leakage node, ρ is the density of the fluid, and γ is the pressure exponent.

[0022] Preferably, the water consumption calculation formula for the DMA in step S1.4 satisfies:

[0023] Q i =h i (Q in -Q out )

[0024] Among them, Q iis the water consumption of the ith metering zone, h i is the leakage duration of the i-th metering partition, Q in is the flow rate into the metering partition per unit time, Q out It is the flow rate outflowing from the metering partition per unit time.

[0025] Preferably, in step S4, a leakage location model is established with the minimum square error between the monitoring value and the simulation value of the pressure measuring point when leakage occurs as the objective function, wherein the objective function is:

[0026]

[0027] Among them, N s is the number of pressure measuring points, T is the time series, p 1,s (t), p 2,s (t) are the simulation value and monitoring value of the pressure measuring point respectively.

[0028] Compared with the prior art, the leakage monitoring method of a town water supply network provided by the present invention has the following beneficial effects: The present invention provides a leakage monitoring method of a town water supply network, the method comprising:

[0029] Step S1: construct a dynamic water supply network model, and simulate and obtain the leakage flow and the water consumption of the partition metering DMA under the leakage condition based on the constructed dynamic water supply network model; Step S2: construct a neural network water consumption prediction model, the neural network water consumption prediction model uses historical water consumption data, weather conditions and holiday information as input parameters, and predicts the normal water consumption when no leakage occurs through training and learning; Step S3: construct a neural network leakage monitoring model, the neural network leakage monitoring model uses the normal water consumption output in step S2 and the water consumption of the partition metering DMA under the leakage condition as input parameters to determine whether each DMA partition has leakage; Step S4: for the DMA partition where leakage occurs, a leakage positioning model is established with the minimum square error between the monitoring value and the simulation value of the pressure measuring point when the leakage occurs as the objective function, and the leakage positioning model is solved to locate the leakage position of the DMA partition. The present invention predicts the normal water consumption through the neural network model, locates the leakage of the DMA partition where leakage occurs, determines the leakage position by solving the leakage positioning model, reduces the dependence on additional equipment and data, and improves the estimation accuracy.

[0030] The present invention also provides a system for monitoring leakage of urban water supply pipe networks, the system comprising:

[0031] The water supply network model module is used to construct a dynamic water supply network model, and simulate the leakage flow and the water consumption of the DMA under leakage conditions based on the constructed dynamic water supply network model;

[0032] A water consumption prediction model is used to construct a neural network water consumption prediction model, which uses historical water consumption data, weather conditions and holiday information as input parameters, and predicts normal water consumption when no leakage occurs through training and learning;

[0033] A leakage monitoring model is used to construct a neural network leakage monitoring model, wherein the neural network leakage monitoring model uses the normal water consumption output in step S2 and the water consumption of the partition metering DMA under the leakage condition as input parameters to determine whether leakage occurs in each DMA partition;

[0034] The leakage location module is used to establish a leakage location model for the DMA partition where leakage occurs, taking the minimum square error between the monitoring value and the simulation value of the pressure measuring point when leakage occurs as the objective function, and solve the leakage location model to locate the DMA.

[0035] Partition leakage location.

[0036] Preferably, the water supply network model module includes:

[0037] Water supply network model unit, used to build a dynamic water supply network model using EPANET software model;

[0038] A simulation unit is used to set up diffusers on EPANET software model nodes to simulate leakage accidents;

[0039] The leakage flow unit is used to call the dynamic link library of the EPANET software model to calculate the leakage flow;

[0040] The water consumption unit is used to set up metering water meters on the water supply pipes at the boundaries of each DMA partition in the dynamic water supply network model to simulate the actual water metering process. According to the water balance principle, the water consumption of the partition metering DMA under leakage conditions is obtained by calculating the difference in water inflow and outflow in each DMA partition.

[0041] Preferably, the calculation formula of the leakage flow satisfies:

[0042]

[0043] Among them, q is the leakage flow through the diffuser, μ is the flow coefficient, which is used to characterize the severity of pipeline leakage, p is the pressure at the leakage node, ρ is the density of the fluid, and γ is the pressure exponent.

[0044] Preferably, the water consumption calculation formula of the partition metering DMA in the water consumption unit satisfies:

[0045] Q i =h i (Q in -Q out )

[0046] Among them, Q i is the water consumption of the ith metering zone, h i is the leakage duration of the i-th metering partition, Q in is the flow rate into the metering partition per unit time, Q out It is the flow rate outflowing from the metering partition per unit time.

[0047] Preferably, in the leakage locating module, a leakage locating model is established with the minimum square error between the monitoring value and the simulation value of the pressure measuring point when leakage occurs as the objective function, wherein the objective function is:

[0048]

[0049] Among them, N s is the number of pressure measuring points, T is the time series, p 1,s (t), p 2,s (t) are the simulation value and monitoring value of the pressure measuring point respectively.

[0050] Compared with the prior art, the beneficial effects of the urban water supply network leakage monitoring system provided by the present invention are the same as the beneficial effects of the urban water supply network leakage monitoring method described in the above technical solution, which will not be repeated here.

[0051] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0053] Figure 1 A flow chart of a method for monitoring leakage in a town water supply network provided by an embodiment of the present invention is shown;

[0054] Figure 2 A schematic diagram of a water supply network model provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0055] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0056] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0057] The embodiment of the present invention provides a method for monitoring leakage of a town water supply network. Figure 1 A flow chart of a method for monitoring leakage of a town water supply network provided by an embodiment of the present invention is shown. Figure 1 As shown, the method includes:

[0058] Step S1: construct a dynamic water supply network model, and simulate and obtain the leakage flow and the water consumption of the zone metering DMA under leakage conditions based on the constructed dynamic water supply network model.

[0059] It should be noted that the step S1 includes:

[0060] Step S1.1: Use the EPANET software model to build a dynamic water supply network model. EPANET is an open source water distribution system hydraulic and water quality simulation software developed by the US Environmental Protection Agency. The water distribution system in EPANET consists of pipe segments and nodes. Pipe segments represent components such as pipelines, pumps, and valves. Nodes represent components such as water use nodes, connection nodes, pools, and reservoirs. Attributes can be set for each component, such as the length of the pipeline, the elevation of the node, the flow and head of the water pump, etc. By creating connection nodes and pipe segments and setting their attributes, a water supply network model can be described and established. In order to obtain flow data under leakage accidents, the following is constructed in EPANET: Figure 2 The water supply network model is shown.

[0061] Step S1.2: Set up a diffuser on the EPANET software model node to simulate a leakage accident.

[0062] In the pipe network model, one or more nodes are selected as the locations for leakage simulation according to actual needs and possible leakage points. These nodes are usually key nodes in the pipe network, such as pipe intersections, old pipe sections, etc. In the EPANET model, there is no direct "diffuser" component, but leakage can be simulated by setting specific node properties. The present invention is achieved by setting a "virtual water point" with a large flow demand on the selected node, and can also simulate the energy loss caused by leakage by modifying the head loss coefficient of the node. Add a virtual water point to the selected node and set a large flow value, which represents the flow of the leakage. It should be understood that at this time, it is necessary to ensure that the flow of the virtual water point will not have too much impact on the hydraulic balance of the entire pipe network. The head loss coefficient of the selected node can also be modified to be greater than the value under normal circumstances. In this way, the energy loss caused by leakage can be simulated, thereby affecting the water pressure and flow distribution in the pipe network, so as to more accurately simulate the impact of leakage on the pipe network. Finally, according to the severity and characteristics of the leakage, adjust the flow value of the virtual water point or the head loss coefficient of the node,

[0063] Step S1.3: Call the dynamic link library of the EPANET software model to calculate the leakage flow.

[0064] First, the dynamic link library DLL of the EPANET software model is called. Before calling the DLL, the EPANET software model must first be initialized. According to the leakage simulation conditions set in step S1.2, the flow value of the virtual water use point or the head loss coefficient of the node is adjusted. The present invention takes the adjustment of the flow value of the virtual water use point as an example to perform calculations. The calculation formula of the leakage flow satisfies:

[0065]

[0066] Among them, q is the leakage flow through the diffuser, Q is the flow value of the virtual water point, μ is the flow coefficient, which is used to characterize the severity of pipeline leakage, p is the pressure at the leakage node, ρ is the density of the fluid, and γ is the pressure index. When the flow coefficient is 0, there is no leakage at the node. When the flow coefficient is greater than 0, there is leakage at the node, and the larger the flow coefficient, the larger the leakage flow. It should be noted that since the pressure at the node will decrease with the increase of the leakage flow, the leakage flow is not proportional to the flow coefficient.

[0067] Step S1.4: In the dynamic water supply network model, metering water meters are set on the water supply pipes at the boundaries of each DMA partition to simulate the actual water metering process. According to the principle of water balance, the water volume difference between the inflow and outflow of each DMA partition is calculated to obtain the water consumption of the partition metering DMA under the leakage condition.

[0068] When leakage occurs in a metering zone, the most direct change is an abnormal increase in water consumption. The water consumption of the metering zone is obtained through the metering water meter installed on the water supply pipeline. According to the water balance principle, the water consumption calculation formula of the zone metering DMA satisfies:

[0069] Q i =h i (Q in -Q out )

[0070] Among them, Q i is the water consumption of the ith metering zone, h i is the leakage duration of the i-th metering partition, Q in is the flow rate into the metering partition per unit time, Q out It is the flow rate outflowing from the metering partition per unit time.

[0071] Step S2: Construct a neural network water consumption prediction model, which uses historical water consumption data, weather conditions and holiday information as input parameters, and predicts normal water consumption when no leakage occurs through training and learning.

[0072] It should be noted that, first, the historical water consumption data in the past three months are collected and sorted. These data contain water consumption under different seasons, weather and holiday conditions to fully reflect the changing pattern of water consumption. At the same time, weather condition data such as temperature, humidity, rainfall, etc., as well as holiday information, including the date and type of holidays, are obtained in the corresponding time period. Then, the collected data is preprocessed, including data cleaning, such as removing missing values ​​and outliers, extracting features useful for water consumption prediction from the original data, and performing data standardization / normalization. Then, based on the preprocessed data, a neural network model is constructed. The model adopts a multi-layer perceptron (MLP) network architecture to fully capture the time series characteristics in water consumption data and the complex relationship between other influencing factors. The input layer of the model will accept historical water consumption data, weather conditions and holiday information as input parameters, and through the processing of the hidden layer, the output layer will finally give the normal water consumption forecast value for a period of time in the future. In the model training stage, a part of the data is used as a training set to train the neural network, and the error between the predicted value and the actual value is minimized by continuously adjusting the network weights and bias terms. At the same time, cross-validation and other techniques are used to evaluate the performance of the model and prevent overfitting. After training is completed, the remaining data is used as a test set to evaluate the prediction accuracy and generalization ability of the model. If the model performs well, it can be applied to actual scenarios to predict future normal water consumption based on real-time and historical weather conditions, holiday information, and possible other influencing factors.

[0073] Step S3: construct a neural network leakage monitoring model, which uses the normal water consumption output in step S2 and the water consumption of the partition metering DMA under leakage conditions as input parameters to determine whether leakage occurs in each DMA partition.

[0074] It should be noted that the predicted value of normal water consumption obtained in step S2 and the water consumption under leakage condition simulated in step S1 are used as input parameters of the neural network leakage monitoring model. At the same time, the logical value of whether leakage occurs in each DMA partition is defined, 0 means no leakage occurs, 1 means leakage occurs, as the output parameter of the model. Through a large amount of learning and training, the neural network learns the difference pattern between normal water consumption and water consumption under leakage condition, thereby identifying abnormal data and inferring whether leakage occurs in each DMA partition.

[0075] The two neural networks of step S2 and step S3 can also be combined into one, and the five water consumption influencing factors such as water use time, holidays, daily maximum temperature, daily minimum temperature, weather type, and the partitioned DMA water consumption under leakage conditions can be directly used as input parameters, and whether leakage occurs in each partition can be used as the output parameter. In this way, leakage diagnosis of multiple DMA partitions can be achieved through one neural network.

[0076] Step S4: For the DMA partition where leakage occurs, a leakage location model is established with the minimum square error between the monitored value and the simulated value of the pressure measuring point when leakage occurs as the objective function, and the leakage location model is solved to locate the leakage position of the DMA partition.

[0077] It should be noted that the pressure and flow of the pipe network operating under normal conditions are in a dynamic equilibrium steady state. When any pipe in the water supply pipe network leaks, it will destroy the original steady state in the pipe network. At this time, a certain amount of injection flow will be generated at the leakage position. There is a positive correlation between flow and pressure, which can be specifically expressed as:

[0078] Q i (t) = k i [p i (t)] n

[0079] Among them, Q i (t) is the total outflow of node i at time t, p i (t) is the pressure of node i at time t, k i is the leakage coefficient, n is the leakage index, and the value ranges from 0.5 to 2.5.

[0080] The pressure fluctuation of the pipeline network during leakage mainly depends on the location of the leakage and the size of the leakage flow. Therefore, if the changes in pressure data during abnormal pipeline network are known, the leakage location and the amount of water loss can be reversely calculated using the pipeline network hydraulic model and the pipeline network leakage model. The leakage location model is established with the minimum square error between the monitoring value and the simulation value of the pressure measuring point when the leakage occurs as the objective function, where the objective function is:

[0081]

[0082] Among them, N s is the number of pressure measuring points, T is the time series, p 1,s (t), p 2,s (t) are the simulation value and monitoring value of the pressure measuring point respectively.

[0083] The leakage location is regarded as a variable to be solved, and possible leakage points in the pipe network are represented by parameters. Each candidate leakage point is associated with a potential leakage flow, and these parameters together constitute the input of the leakage location model. Using the pipe network hydraulic model, each candidate leakage location and the assumed leakage flow are simulated to generate the corresponding pipe network pressure distribution data. The particle swarm optimization algorithm is used to solve the leakage location model. By iteratively adjusting the values ​​of the leakage location and leakage flow, the objective function is continuously optimized until the optimal solution with the minimum square error is found, so as to screen out the most likely leakage location from the candidate leakage point set.

[0084] Compared with the prior art, the leakage monitoring method for urban water supply network provided by the embodiment of the present invention has the following beneficial effects: through training and learning, the normal water consumption when no leakage occurs is predicted;

[0085] Step S3: constructing a neural network leakage monitoring model, wherein the neural network leakage monitoring model uses the normal water consumption output in step S2 and the water consumption of the DMA under leakage conditions as input parameters to determine whether leakage occurs in each DMA partition;

[0086] The present invention predicts normal water consumption through a neural network water consumption prediction model, takes the normal water consumption and the water consumption of the DMA under leakage conditions as input parameters of the neural network leakage monitoring model, and judges whether leakage occurs in each DMA partition.

[0087] For DMA partitions where leakage occurs, a leakage location model is established by taking the minimum square error between the monitored value and the simulated value at the pressure measuring point when leakage occurs as the objective function. The leakage location is determined by solving the leakage location model, which reduces dependence on additional equipment and data and improves estimation accuracy.

[0088] The embodiment of the present invention further provides a system for monitoring leakage of urban water supply pipe network, the system comprising:

[0089] The water supply network model module is used to construct a dynamic water supply network model, and simulate the leakage flow and the water consumption of the DMA under leakage conditions based on the constructed dynamic water supply network model;

[0090] A water consumption prediction model is used to construct a neural network water consumption prediction model, which uses historical water consumption data, weather conditions and holiday information as input parameters, and predicts normal water consumption when no leakage occurs through training and learning;

[0091] A leakage monitoring model is used to construct a neural network leakage monitoring model, wherein the neural network leakage monitoring model uses the normal water consumption output in step S2 and the water consumption of the partition metering DMA under the leakage condition as input parameters to determine whether leakage occurs in each DMA partition;

[0092] The leakage location module is used to establish a leakage location model for the DMA partition where leakage occurs, taking the minimum square error between the monitoring value and the simulation value of the pressure measuring point when leakage occurs as the objective function, and solve the leakage location model to locate the DMA.

[0093] Partition leakage location.

[0094] Preferably, the water supply network model module includes:

[0095] Water supply network model unit, used to build a dynamic water supply network model using EPANET software model;

[0096] A simulation unit is used to set up diffusers on EPANET software model nodes to simulate leakage accidents;

[0097] The leakage flow unit is used to call the dynamic link library of the EPANET software model to calculate the leakage flow;

[0098] The water consumption unit is used to set up metering water meters on the water supply pipes at the boundaries of each DMA partition in the dynamic water supply network model to simulate the actual water metering process. According to the water balance principle, the water consumption of the partition metering DMA under leakage conditions is obtained by calculating the difference in water inflow and outflow in each DMA partition.

[0099] Preferably, the calculation formula of the leakage flow satisfies:

[0100]

[0101] Among them, q is the leakage flow through the diffuser, μ is the flow coefficient, which is used to characterize the severity of pipeline leakage, p is the pressure at the leakage node, ρ is the density of the fluid, and γ is the pressure exponent.

[0102] Preferably, the water consumption calculation formula of the partition metering DMA in the water consumption unit satisfies:

[0103] Q i =h i (Q in -Q out )

[0104] Among them, Q i is the water consumption of the ith metering zone, h i is the leakage duration of the i-th metering partition, Q in is the flow rate into the metering partition per unit time, Q out It is the flow rate outflowing from the metering partition per unit time.

[0105] Preferably, in the leakage locating module, a leakage locating model is established with the minimum square error between the monitoring value and the simulation value of the pressure measuring point when leakage occurs as the objective function, wherein the objective function is:

[0106]

[0107] Among them, N s is the number of pressure measuring points, T is the time series, p 1,s (t), p 2,s (t) are the simulation value and monitoring value of the pressure measuring point respectively.

[0108] Compared with the prior art, the beneficial effects of a system for monitoring leakage of a urban water supply network provided by an embodiment of the present invention are the same as the beneficial effects of a method for monitoring leakage of a urban water supply network described in the above technical solution, which will not be elaborated here.

[0109] In addition, an embodiment of the present invention also provides an electronic device, including a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and run on the processor. The transceiver, the memory and the processor are respectively connected via a bus. When the computer program is executed by the processor, each process of the above-mentioned embodiment of a method for monitoring leakage in a town water supply network is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0110] In addition, an embodiment of the present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the various processes of the above-mentioned urban water supply network leakage monitoring method embodiment are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0111] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technical solution that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for monitoring leakage of urban water supply pipe network, Its characteristics are, include: Step S1: construct a dynamic water supply network model, and simulate and obtain the leakage flow and the water consumption of the DMA under leakage conditions based on the constructed dynamic water supply network model; Step S2: constructing a neural network water consumption prediction model, wherein the neural network water consumption prediction model uses historical water consumption data, weather conditions and holiday information as input parameters, and predicts normal water consumption when no leakage occurs through training and learning; Step S3: constructing a neural network leakage monitoring model, wherein the neural network leakage monitoring model uses the normal water consumption output in step S2 and the water consumption of the DMA under leakage conditions as input parameters to determine whether leakage occurs in each DMA partition; Step S4: For the DMA partition where leakage occurs, a leakage location model is established with the minimum square error between the monitored value and the simulated value of the pressure measuring point when leakage occurs as the objective function, and the leakage location model is solved to locate the leakage position of the DMA partition.

2. A method for monitoring leakage of urban water supply pipe network according to claim 1, It is characterized in that The step S1 comprises: Step S1.1: Use EPANET software model to build a dynamic water supply network model; Step S1.2: Setting up a diffuser on the EPANET software model node to simulate a leakage accident; Step S1.3: calling the dynamic link library of the EPANET software model to calculate the leakage flow; Step S1.4: In the dynamic water supply network model, metering water meters are set on the water supply pipes at the boundaries of each DMA partition to simulate the actual water metering process. According to the principle of water balance, the water volume difference between the inflow and outflow of each DMA partition is calculated to obtain the water consumption of the partition metering DMA under the leakage condition.

3. A method for monitoring leakage of urban water supply pipe network according to claim 2, It is characterized in that The calculation formula of the leakage flow in step S1.3 satisfies: Among them, q is the leakage flow through the diffuser, μ is the flow coefficient, which is used to characterize the severity of pipeline leakage, p is the pressure at the leakage node, ρ is the density of the fluid, and γ is the pressure exponent.

4. A method for monitoring leakage of urban water supply pipe network according to claim 2, It is characterized in that The water consumption calculation formula of the DMA in step S1.4 satisfies: Q i =h i (Q in -Q out ) Among them, Q i is the water consumption of the ith metering zone, h i is the leakage duration of the i-th metering partition, Q in is the flow rate into the metering partition per unit time, Q out It is the flow rate outflowing from the metering partition per unit time.

5. A method for monitoring leakage of urban water supply pipe network according to claim 1, It is characterized in that In step S4, a leakage location model is established with the minimum square error between the monitoring value and the simulation value of the pressure measuring point when leakage occurs as the objective function, wherein the objective function is: Among them, N s is the number of pressure measuring points, T is the time series, p 1,s (t), p 2,s (t) are the simulation value and monitoring value of the pressure measuring point respectively.

6. A leakage monitoring system for urban water supply network, It is characterized in that include: The water supply network model module is used to construct a dynamic water supply network model, and simulate the leakage flow and the water consumption of the DMA under leakage conditions based on the constructed dynamic water supply network model; A water consumption prediction model is used to construct a neural network water consumption prediction model, which uses historical water consumption data, weather conditions and holiday information as input parameters, and predicts normal water consumption when no leakage occurs through training and learning; A leakage monitoring model is used to construct a neural network leakage monitoring model, wherein the neural network leakage monitoring model uses the normal water consumption output in step S2 and the water consumption of the partition metering DMA under the leakage condition as input parameters to determine whether leakage occurs in each DMA partition; The leakage locating module is used to establish a leakage locating model for the DMA partition where leakage occurs, taking the minimum square error between the monitoring value and the simulation value of the pressure measuring point when leakage occurs as the objective function, and solve the leakage locating model to locate the leakage position of the DMA partition.

7. A system for monitoring leakage of urban water supply pipe network according to claim 6, It is characterized in that The water supply network model module includes: Water supply network model unit, used to build a dynamic water supply network model using EPANET software model; A simulation unit is used to set up diffusers on EPANET software model nodes to simulate leakage accidents; The leakage flow unit is used to call the dynamic link library of the EPANET software model to calculate the leakage flow; The water consumption unit is used to set up metering water meters on the water supply pipes at the boundaries of each DMA partition in the dynamic water supply network model to simulate the actual water metering process. According to the water balance principle, the water consumption of the partition metering DMA under leakage conditions is obtained by calculating the difference in water inflow and outflow in each DMA partition.

8. A system for monitoring leakage of urban water supply pipe network according to claim 7, It is characterized in that The calculation formula of leakage flow satisfies: Among them, q is the leakage flow through the diffuser, μ is the flow coefficient, which is used to characterize the severity of pipeline leakage, p is the pressure at the leakage node, ρ is the density of the fluid, and γ is the pressure exponent.

9. A system for monitoring leakage of urban water supply pipe network according to claim 7, It is characterized in that The water consumption calculation formula of the DMA in the water consumption unit satisfies: Q i =h i (Q in -Q out ) Among them, Q i is the water consumption of the ith metering zone, h i is the leakage duration of the i-th metering partition, Q in is the flow rate into the metering partition per unit time, Q out It is the flow rate outflowing from the metering partition per unit time.

10. A system for monitoring leakage of urban water supply pipe network according to claim 6, It is characterized in that In the leakage locating module, a leakage locating model is established with the minimum square error between the monitoring value and the simulation value of the pressure measuring point when leakage occurs as the objective function, wherein the objective function is: Among them, N s is the number of pressure measuring points, T is the time series, p 1,s (t), p 2,s (t) are the simulation value and monitoring value of the pressure measuring point respectively.

Images