A water pipe network leakage monitoring point layout optimization method

Abstract

The application provides a water pipe network leakage monitoring point layout optimization method. The method comprises the following steps: (1) obtaining pipe network information; (2) carrying out digital pretreatment on the pipe network information; (3) carrying out initial layout on the monitoring points; (4) determining the optimal number of monitoring points; and (5) obtaining the optimal layout of the monitoring points. The application fully utilizes the propagation characteristics and attenuation characteristics of fluid wave signals, establishes a calculation model of signal amplitude and fluid wave signal propagation distance, realizes quantitative evaluation of the attenuation of fluid wave signals of the leakage points, and sets the shortest interval between the monitoring points according to the evaluation, and the reliability of the result is high. The application combines the leakage monitoring point layout optimization and the fluid wave characteristics, and has the characteristics of high acoustic signal sensitivity, small error, wide detection range and the advancement of the genetic algorithm.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of water pipe network leakage monitoring, and particularly relates to a water pipe network leakage monitoring point layout optimization method. BACKGROUND

[0002] Among the industries with large water consumption in China, the petrochemical industry is one of them. However, due to the current shortage of water resources, the development of this industry has been seriously affected. Water conservation is related to the healthy development of various industries, and more importantly, it is related to the future of the country. In recent years, water leakage frequently occurs in water pipelines in refining enterprises, causing great waste of water resources and huge economic losses, affecting the normal production and safe operation of enterprises. Considering the large amount of work, high technical requirements for personnel, time lag and other problems of on-site leakage detection, according to the principle of fluid dynamics, sensors such as acceleration sensors and vibration sensors are used to collect leakage signals, and the signals at the pipe leakage point are analyzed to obtain the pressure fluctuation at the pipe leakage point, which is the source of pipe vibration. According to the wave speed, it is concluded that the signal belongs to the fluid wave signal, and it is analyzed that the fluid wave attenuation is mainly caused by the viscous absorption of the medium.

[0003] In the research of fluid waves, acoustic methods are often used to detect pipeline leaks. Compared with flow method, pressure method, chemical method and stress wave method, it has the advantages of high sensitivity, small error and wide detection frequency range. Genetic algorithm is used to optimize the original positioning formula of the water supply pipeline leakage point, and a more accurate optimization positioning model is obtained, so as to improve the accuracy of water supply pipeline leakage point positioning. In the field of fluid wave research, there is a lack of work on the optimization of monitoring point layout, which mainly focuses on improving the positioning accuracy and reducing the error.

[0004] At present, in the research of positioning the leakage point of water supply pipe network based on fluid pressure attenuation, the traditional negative pressure wave algorithm based on time difference method is improved, and the ANPW algorithm based on pressure attenuation is proposed. The attenuation characteristics of negative pressure wave are used, and the accuracy of ANPW algorithm in leakage positioning is improved through denoising processing, so that ANPW algorithm can play a good effect in water supply pipeline positioning.

[0005] In the field of sensor optimization arrangement of water supply pipe network leakage monitoring, clustering analysis method is currently used. Since the clustering analysis method is sensitive to k value (the number of categories is k), the selection of k has a great influence on the result. According to the existing data, if the data volume is large, the error caused by classification will also be large.

[0006] At present, in the field of city water supply pipe network burst prediction, due to the uncertainty of water supply pipe network burst itself, the variability of influencing factors and the complexity of actual water supply pipe network, the prediction accuracy is not high.

[0007] In the research on monitoring of pipe burst of water supply network, whether pipe burst occurs is determined according to the pressure difference calculated based on the flow rate increase after pipe burst. The number of monitoring points is determined according to marginal effect, and the solving process is ended when the result calculated satisfies the termination condition, i.e. the shortest interval between the monitoring points satisfies the constraint condition. When the constraint condition is not satisfied, the process is repeated, which wastes calculation amount. The monitoring method involves pressure signal, and the pressure fluctuation method is only applicable to the condition of pressure, i.e. water supply pipeline, and is not applicable to complex water pipeline. SUMMARY

[0008] In order to overcome the above-mentioned defects in the prior art, the application provides a water pipeline network leakage monitoring point layout optimization method, which is improved in that the method comprises

[0009] (1) obtaining pipeline network information;

[0010] (2) digitally pre-processing the pipeline network information;

[0011] (3) initially laying the monitoring points;

[0012] (4) determining the optimal number of monitoring points;

[0013] (5) obtaining the optimal layout of the monitoring points.

[0014] Preferably, the step (1) comprises

[0015] For the data preparation step, the signal is collected by the sensor (such as vibration sensor, acceleration sensor, etc.) installed on the outer side of the pipeline wall, the propagation speed of the signal is calculated, and it is determined that the signal is fluid wave. According to the relationship between wave number and wave speed, the prediction formula of the fluid wave speed c in the leakage pipeline is obtained:

[0016]

[0017] β=2Ba(1-σ 2 ) / (jE) (2)

[0018] k L =f 2 ρ(1-σ 2 ) / E (3)

[0019] In the formula, vf is the wave speed in the fluid medium in an infinite space, β is the bulk modulus of the fluid, σ is the Poisson's ratio of the pipe material, a is the radius of the pipe, kL is the wave number of the plane compression wave, E is the elastic modulus of the pipe wall, B is the bulk modulus of standard water, j is the wall thickness; the nodes are pipe branch points or flow direction changing points, the position information of the pipe and the nodes, the length of the pipe, the diameter of the pipe, and the risk level of the pipe are obtained, and the risk level of the pipe is divided into two levels, wherein the first level is a pipe with higher importance, and the second level is a pipe with lower importance.

[0020] Preferably, the step (2) digitizes and pre-processes the pipe network information, and processes the pipe network data to obtain the following matrices: a pipe section length matrix l, a leakage amount matrix Q, a pipe section weight matrix W, a monitoring point coordinate matrix J, and a monitoring point and pipe section association matrix L.

[0021] Further, the setting rule of the monitoring points is that at least one monitoring point exists every distance d for each pipe to ensure full coverage of the pipe network, and each two monitoring points are called a pipe section, each pipe section is numbered, and if each pipe section leaks, the leakage amount of each pipe section is obtained by the difference between the size of the signal measured by the sensors of the adjacent monitoring points and the set value (affected by the environmental noise), that is, the imbalance of the pipe. The size of the association matrix between the monitoring points and the pipe sections is the number of pipe sections (m) x the number of monitoring points (n), if the i monitoring point is on the j pipe section, then the topology matrix Lij=1, otherwise, Lij=0. The following matrices are obtained: a pipe section length matrix l, a leakage amount (imbalance) matrix Q, a pipe section weight matrix W (determined according to the diameter, flow, and risk level of each pipe section), a monitoring point coordinate matrix J, and a monitoring point and pipe section association matrix L.

[0022] Preferably, the step (3) comprises

[0023] The shortest interval between the monitoring points is constrained, for the pipe sections with large weight values, the coverage rate of the pipe sections after the initial optimization is 100%; and for the case with small weight values, the distance between the two monitoring points should not exceed a certain set value.

[0024] Further, the shortest interval constraint condition is set, and a pipe leakage signal attenuation model is used, and then the attenuation formula of the fluid wave amplitude value is:

[0025] p=p0e -αx (4)

[0026] Wherein, the pipe wall and water flow of water supply pipeline exist viscous force effect, when the fluid wave at the leakage port propagates along the pipeline, the sound vibration energy gradually converts into the heat energy of fluid and pipeline; the fluid wave propagation process needs to overcome viscous force work, thereby the vibration energy gradually reduces, that is, the viscous absorption of medium. The actual fluid has viscosity, which is the main reason causing the fluid wave attenuation. The pipe wall and liquid, liquid layers produce interaction force in the process of fluid flowing in the pipeline, that is, viscous force; wherein, the viscous absorption coefficient can be approximately expressed as

[0027]

[0028] Through formula (1) to formula (5), the relationship between the normalized amplitude of the water supply pipeline leakage signal and the propagation distance is obtained by combining the pipeline condition parameters, the influence law of pipe material, pipe diameter and wall thickness on the attenuation is considered respectively, thereby the fluid wave attenuation model is obtained, and the minimum interval between adjacent monitoring points is obtained according to the calculation result of the model.

[0029] Preferably, the step (4) comprises that the optimized monitoring points satisfy that the minimum interval is not less than a certain set value, a new monitoring point selection range (assuming n) is obtained, the number of monitoring points is taken as a new condition, and the coverage rate of the pipe network under each number of monitoring points is calculated by using the random traversal method from 5 to n in turn. Each calculation result is compared, and finally the maximum value of the coverage rate under the number of monitoring points is output. When the maximum value of the coverage rate under each number is output, when the maximum value of the coverage rate is greater than b with the increase of the number, and the absolute value of the difference between the maximum value and the previous and next maximum values is less than a, the number of monitoring points is output.

[0030] Further, the random traversal method is used when calculating the maximum value of the coverage rate, and the monitoring points provided by the minimum interval optimization are combined according to the size of the number, and part of the combinations are randomly selected, and the specific information in each combination is the monitoring point number which is different from each other. The covered range of each pipe section under the combination is calculated; wherein, the coverage rate is calculated by traversing each pipe section, when the pipe section is monitored by two monitoring points at the same time, when the length of the pipe section is greater than c, the monitored pipe section length is c; when the length of the pipe section is less than c, the monitored pipe section length is the total length of the pipe section; when the pipe section is monitored by one monitoring point, when the length of the pipe section is greater than c / 2, the monitored pipe section length is c / 2; when the length of the pipe section is less than c / 2, the monitored pipe section length is the total length of the pipe section;

[0031] The length of all pipelines is calculated as l2, the number of monitoring points is taken as i, 100x i times are traversed respectively, the length of the pipe section covered by the selected monitoring points each time (the cumulative length is l1) is calculated, and the ratio of the two l1 / l2 is taken as the size of the pipe network coverage rate;

[0032] The maximum coverage rate of each monitoring point number is calculated by comparison method, each calculation result is compared with the maximum value, and finally the maximum value of the coverage rate under the number is output.

[0033] Preferably, the step (5) adopts genetic algorithm to optimize the layout of the monitoring points, sets the fitness as the maximum monitoring leakage, and selects the individual with the maximum fitness value under the number, wherein the output results of the algorithm include: the layout information of the optimized monitoring points, the size of the coverage rate of different numbers, the iteration process of the maximum monitoring leakage, the final optimized number of the monitoring points, the coverage rate and the maximum leakage, and the number of the optimized monitoring points.

[0034] Further, the genetic algorithm is adopted to optimize the layout of the monitoring points, and the optimization of the layout of the monitoring points and the number of the monitoring points is the essence of the optimization of the layout of the leakage monitoring points, which belongs to a nonlinear discrete problem and is solved based on the basic information of the pipe network by using the genetic algorithm.

[0035] The fitness function is constructed, and the maximum monitoring leakage is taken as the fitness function as follows:

[0036]

[0037] In the function, the length l of each pipe section, the unbalance amount AQ of each pipe section, the coordinate information of each monitoring point, the weight coefficient W of each pipe section, and the topological matrix L between the monitoring points and the pipe sections are included.

[0038] The maximum monitoring leakage is taken as the optimization target, the number n1 of the monitoring points is obtained under the constraint condition of the previous coverage rate, the population number is set as 500 times of the number of the monitoring points, the iteration number is set as y times, the fitness value of each individual containing n1 genes (the number of the monitoring points) in the population is calculated, the individual corresponding to the maximum fitness value in the parent population is obtained, genetic selection, crossover, mutation, screening are performed, and after y times of repeated iterations, the gene of the individual with the maximum monitoring leakage under the number is obtained and the related information is output.

[0039] In the operation design, the roulette selection is a playback type random sampling method, and the probability of each individual being left to the next generation is the proportion of the fitness value of the individual to the sum of the fitness values of all individuals. The higher the fitness of the individual, the greater the probability of being selected. Therefore, the roulette selection is adopted.

[0040] The crossover mode is designed, the crossover operation is that two paired chromosomes exchange part of the genes in a certain way according to the crossover probability, so as to form two new individuals. The basic idea of the improved single-point crossover is to exchange the two chromosomes at the crossover position, and the new two chromosomes do not contain the same genes.

[0041] The variation mode is designed, and the gene duplication problem is generated in the variation process, so it is necessary to recheck and avoid the problem from occurring.If the generated gene is duplicated with the existing gene code, it is necessary to re-variation again until the requirement is met.

[0042] The technical scheme provided by the embodiment of the application can include the following beneficial effects:

[0043] The application fully utilizes the propagation characteristics and attenuation characteristics of fluid wave signals, establishes a calculation model of signal amplitude and fluid wave signal propagation distance, realizes quantitative evaluation of fluid wave signal attenuation of the leakage point, and sets the shortest interval between monitoring points, and the result has high reliability.

[0044] The application first performs the shortest interval constraint, limits the distance between monitoring points, saves time and reduces the calculation amount by using the algorithm.

[0045] The application is applied to water pipelines, including water supply pipelines and drainage pipelines, has a wider application range and stronger adaptability.

[0046] Fluid waves are mostly used in leakage detection and positioning research, and the signal characteristics and attenuation characteristics are specifically studied to improve positioning accuracy, the application combines the improved genetic algorithm with fluid wave characteristics, and has high acoustic signal sensitivity, small error, wide detection range and the advancement of the genetic algorithm.

[0047] The application uses the improved genetic algorithm to optimize the layout of monitoring points, does not need to consider the classification of monitoring points, selects individuals by the survival of the fittest, and has strong result reliability.

[0048] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the application. BRIEF DESCRIPTION OF DRAWINGS

[0049] Other features, objects and advantages of the application will become more apparent from the following detailed description of non-limiting embodiments, made with reference to the accompanying drawings:

[0050] Figures 1-3 It is a normalized amplitude and propagation distance relationship diagram of a water pipeline leakage signal in a water pipeline network leakage monitoring point layout optimization method according to the application;

[0051] Figure 4 It is a digitized preprocessing data diagram of one specific embodiment of a water pipeline network leakage monitoring point layout optimization method according to the application;

[0052] Figure 5is a monitoring point layout schematic diagram according to one specific embodiment of a water pipe network leakage monitoring point layout optimization method of the present application;

[0053] Figures 6-7 is monitoring point layout data schematic diagram according to one specific embodiment of a water pipe network leakage monitoring point layout optimization method of the present application;

[0054] Figures 8-9 is monitoring point layout data schematic diagram according to one specific embodiment of a water pipe network leakage monitoring point layout optimization method of the present application;

[0055] Figure 10 is a flow schematic diagram according to one specific embodiment of a water pipe network leakage monitoring point layout optimization method of the present application;

[0056] The same or similar reference signs in the drawings represent the same or similar components. DETAILED DESCRIPTION

[0057] In order to better understand and illustrate the present application, the present application will be further described in detail below in conjunction with the accompanying drawings. The present application is not limited to only these specific embodiments. Rather, modifications or equivalent replacements to the present application shall be encompassed within the scope of the claims of the present application.

[0058] It should be noted that numerous specific details are given in the following specific embodiments. Those skilled in the art should understand that the present application can also be implemented without these specific details. In the following multiple specific embodiments, principles, structures and components well known in the art are not described in detail in order to highlight the main ideas of the present application.

[0059] The present application will be further described below in conjunction with the accompanying drawings and examples:

[0060] The present application utilizes the propagation characteristics and attenuation characteristics of fluid wave signals to obtain the relationship between the distance of the leakage point and the degree of attenuation, and realizes quantitative evaluation of the fluid wave signal attenuation of the leakage point. Unlike other related applications, the present application is applied to the effective distance monitored by the monitoring point, so that the shortest interval between adjacent monitoring points can be set scientifically and reasonably.

[0061] The present application combines the layout optimization of the leakage monitoring point with the fluid wave characteristics, has the characteristics of high acoustic signal sensitivity, small error, wide detection range and the advancement of genetic algorithm. Compared with other related applications in the art, the present application can control the size of the pipe network coverage while limiting the distance between monitoring points, thereby obtaining the optimized number of monitoring points, ensuring that the number of monitoring points is set economically and efficiently. The present application uses an improved genetic algorithm to optimize the layout of the monitoring points, without considering the classification of the monitoring points, and selects individuals through the survival of the fittest. The present application improves the genetic algorithm applied to the layout of the monitoring points, first optimizes in the first step through the shortest interval between adjacent monitoring points, and then solves the coverage rate and the objective function, without repeatedly judging whether the constraint condition (the shortest interval) is met, thereby simplifying the calculation process and reducing the calculation time.

[0062] Example 1

[0063] First step: According to the field data, obtain the field data information required by the algorithm, including the position information of the pipeline and the node, the length of the pipeline, the diameter of the pipeline, the risk level of the pipeline, wherein the risk level of the pipeline is divided into two levels, wherein the first level is the pipeline with higher importance, and the second level is the pipeline with lower importance.

[0064] Second step: As shown in Figures 1-3 , the pipe network information is digitally preprocessed, wherein the setting rule of the monitoring point is that at least one monitoring point exists every distance d of each pipeline to ensure full coverage of the pipe network, and each pipe segment is numbered, assuming that each pipe segment leaks, then the leakage of each pipe segment is the difference between the size of the signal measured by the adjacent monitoring point and the set value, that is, the unbalance of the pipeline. The pipe network data is processed to obtain the following matrix: pipe segment length matrix l, leakage (unbalance) matrix Q, pipe segment weight matrix W (wherein the weight of the relatively important pipeline is 0.8, and the weight of the less important pipeline is 0.5), monitoring point coordinate matrix J, and monitoring point and pipe segment association matrix L, when i monitoring point is on j pipe segment, then Lij=1, otherwise, Lij=0. As shown in Figure 4 , according to the above information, all monitoring point layout diagrams as shown in Figure 5 can be obtained.

[0065] Step 3: In order to prevent the monitoring points from concentrating on some pipes and thus causing the overlap of monitoring areas, the shortest interval between the monitoring points needs to be constrained. For the case where the weight value of the pipe segment is large (0.8 in this case), it is ensured that the coverage of these pipe segments is 100% after optimization; for the case where the weight value is small (0.6 in this case), the distance between two monitoring points should not exceed a certain set value. According to the attenuation model of fluid waves, if the wall thickness of the plastic pipe is 4.5 mm and the diameter of the pipe is 300 mm, according to the attenuation model in Figure 1 , Figure 2 , Figure 3 , when the distance between the monitoring point and the leakage point is 60 m and the amplitude attenuation reaches 85%, the set value is set to 100 m and 120 m respectively, and the results are shown in Figure 6 , Figure 7 .

[0066] Step 4: After the optimization of the monitoring points in the previous step, the shortest interval between the monitoring points is not less than a certain set value, and the new monitoring point selection range (assuming there are n) can be obtained. The number of monitoring points is taken as a new condition, and a specified number of monitoring points is selected from the n, assuming that the number of monitoring points is i, then 100 x i times are traversed respectively. The coverage of the pipe network is calculated from 5 to n monitoring points by using the random traversal method, and each calculation result is compared with the maximum value. Finally, the maximum coverage value under this number is output. The maximum coverage matrix is output, which contains the maximum value of each combination coverage under each number of monitoring points (from 5 to n). When the maximum coverage value increases with the number of points and the absolute value of the difference between the maximum values before and after is less than 0.05, the number of monitoring points under this condition is output as the optimized number of monitoring points.

[0067] Step 5: The fitness is set to the maximum monitoring leakage, as shown in equation (6), and the maximum monitoring leakage is taken as the optimization goal to obtain the number of monitoring points n1 under the constraint condition of the previous coverage. The population size is set to 500 times the number of monitoring points, and the iteration number is set to 100. The fitness value of each individual in the population containing n1 genes (monitoring point number) is calculated, and the individual corresponding to the maximum fitness value in the parent population is obtained. Then genetic selection, crossover, mutation, and screening are performed, and after 100 iterations, the individual gene condition of the maximum monitoring leakage under this number is obtained and the final result is output, which includes: the optimized monitoring point layout information, the size of the coverage under different numbers, the iteration process of the maximum leakage, the final optimized number of monitoring points, the coverage, and the maximum leakage, and the optimized monitoring point number. When the shortest distance is set to 100 m, the final result output is shown in Figure 9 .

[0068] Example 2:

[0069] 1) The influence of pipe material, pipe diameter and wall thickness should be considered when calculating the attenuation characteristics of fluid waves. PVC pipe, cast iron pipe and steel pipe were used respectively, and the pipe parameters and signal frequencies are shown in the following table:

[0070]

[0071]

[0072] 2) Assuming the wall thickness is 4.5 mm and the pipe diameter is 300 mm, the relationship between the normalized amplitude and the signal propagation distance is obtained by calculating formula (1) to formula (5), as shown in Figure 1 .

[0073] 3) Taking 4.5 mm PVC pipe as an example, four pipe diameters, DN30, DN100, DN300 and DN1000, were used. The relationship between the normalized amplitude and the signal propagation distance is obtained by calculating formula (1) to formula (5), as shown in Figure 2 .

[0074] 4) Taking DN300 PVC pipe as an example, three wall thicknesses, 3.5 mm, 4.5 mm and 5.5 mm, were used. The relationship between the normalized amplitude and the signal propagation distance is obtained by calculating formula (1) to formula (5), as shown in Figure 3 .

[0075] 5) According to the above information, if the wall thickness of the PVC pipe is 3.5 mm and the pipe diameter is 300 mm, according to the attenuation model in Figure 1 , Figure 2 , Figure 3 , when the distance between the monitoring point and the leakage point is 60 m and the amplitude attenuation reaches 80%, the set values are set to 100 m and 120 m respectively, and the results are shown in Figure 6 , Figure 7 .

[0076] It should be noted that although the operations of the method of the present application are described in a specific order in the accompanying drawings, this does not require or imply that the operations must be performed in that specific order, or that all of the shown operations must be performed to achieve the desired result. On the contrary, the steps depicted in the flowchart can change the order of execution. Additionally or alternatively, some steps can be omitted, combined into one step, and / or divided into multiple steps.

[0077] Those skilled in the art will understand that any computer system with suitable programming means is capable of executing the steps of the methods of this application contained in a computer program product. Although most of the specific embodiments described in this specification focus on software programs, alternative embodiments that implement the methods provided in this application in hardware are also within the scope of protection claimed in this application.

[0078] It will be apparent to those skilled in the art that this application is not limited to the details of the exemplary embodiments described above, and that this application can be implemented in other specific forms without departing from the spirit or essential characteristics of this application. Therefore, the embodiments should be considered exemplary and not restrictive, and the scope of this application is defined by the appended claims rather than the foregoing description. All variations within the meaning and scope of equivalents of the claims are included within this application. No reference numerals in the claims should be construed as limiting the scope of the claims. Furthermore, it is clear that the word "comprising" does not exclude other components, units, or steps, and the singular does not exclude the plural. Multiple components, units, or devices recited in the claims may also be implemented by a single component, unit, or device through software or hardware.

[0079] The above-disclosed embodiments or specific implementations are only a part of the embodiments or specific implementations of this application and should not be construed as limiting the scope of the rights of this application. Equivalent changes made in accordance with the claims of this application shall still fall within the scope of this application.

Claims

1. A water pipe network leakage monitoring point layout optimization method, characterized in that, The method comprises (1) obtaining pipe network information; (2) digitizing and preprocessing the pipe network information; (3) initially arranging monitoring points; (4) determining the optimal number of monitoring points; (5) obtaining the optimal arrangement of monitoring points; The step (1) comprises: obtaining the position information of the pipeline and nodes, the length of the pipeline, the diameter of the pipeline, and the risk level of the pipeline, the node being a pipeline branch point or a flow direction changing point; a signal is collected by a sensor installed on the outside of the pipe wall, the propagation speed of the signal is calculated, and a prediction formula of the fluid wave speed c in the leaking pipeline is obtained according to the relationship between the wave number and the wave speed: In the formula, vf is the wave speed in the fluid medium in an infinite space, β is the bulk modulus of the fluid, σ is the Poisson's ratio of the pipeline material, a is the radius of the pipeline, kL is the wave number of the plane compression wave, E is the elastic modulus of the pipe wall, B is the bulk modulus of standard water, and j is the wall thickness; The step (3) comprises: The shortest interval between monitoring points is constrained, for a pipeline segment with a large weight value, the coverage rate of the pipeline segment after initial optimization is set to 100%; for a small weight value, the distance between two monitoring points should not exceed a certain set value; The shortest interval constraint condition is set, and a pipeline leakage signal attenuation model is used, so the attenuation formula of the fluid wave amplitude is: （4） There is a viscous force between the pipe wall and the water flow of the water supply pipeline, when the fluid wave at the leakage port propagates along the pipeline, the acoustic vibration energy is gradually converted into heat energy of the fluid and the pipeline; the fluid wave propagation process needs to overcome viscous force work, that is, the viscous absorption of the medium; the viscous absorption coefficient is represented as: （5） Through formulas (1)-(5), the relationship between the normalized amplitude of the water pipeline leakage signal and the propagation distance is obtained by combining the pipeline condition parameters, and the minimum interval between adjacent monitoring points is obtained according to the calculation results of the model; The step (4) comprises that the monitoring points after optimization satisfy that the shortest interval is not less than a certain set value, a new monitoring point selection range is obtained, the number of monitoring points is selected as a new condition, and the coverage rate of the pipeline network under each number of monitoring points is calculated by using a random traversal method from 5 to n; each calculation result is compared, and the maximum coverage rate under the number of monitoring points is output; the maximum coverage rate under each number is output, when the maximum coverage rate increases with the number and is greater than b, and the absolute value of the difference between the maximum values before and after is less than a, the number of monitoring points is output; The step (5) uses a genetic algorithm to optimize the arrangement of monitoring points, and sets the fitness as the maximum monitoring leakage amount, selects the individual with the maximum fitness value under the number, and the output results of the algorithm comprise: the arrangement information of the monitoring points after optimization, the size of the coverage rate of different numbers, the iteration process of the maximum monitoring leakage amount, the final optimized number of monitoring points, the coverage rate and the maximum leakage amount, and the number of monitoring points after optimization; The genetic algorithm is used to optimize the arrangement of monitoring points, the site selection and the number of monitoring points are optimized, and the genetic algorithm is used to solve based on the basic information of the pipeline network: A fitness function is constructed, and the maximum monitoring leakage amount is used as the fitness function as follows: （6） Wherein, the function includes the length of each pipe segment l, the imbalance of each pipe segment ΔQ, the coordinate information of each monitoring point, the weight coefficient of each pipe segment W, and the topological matrix L between the monitoring points and the pipe segments; The maximum monitoring leakage amount is taken as an optimization target, and the number of monitoring points n1 is obtained under the constraint condition of the previous coverage rate. The population number is set to 500 times the number of monitoring points, and the iteration number is set to y times. The fitness value of each individual containing n1 genes in the population is calculated, and the individual corresponding to the maximum fitness value in the parent population is obtained. Genetic selection, crossover, mutation, screening, and repeated iteration y times are performed to obtain the individual gene condition of the maximum monitoring leakage amount under the number and output the related information.

2. The method according to claim 1, wherein, The step (2) processes the pipe network data to obtain a pipe segment length matrix l, a leakage amount matrix Q, a pipe segment weight matrix W, a monitoring point coordinate matrix J, and an associated matrix L of monitoring points and pipe segments.

3. The method according to claim 2, wherein, The setting rule of the monitoring points is that at least one monitoring point exists every distance d for each pipe, and each pipe segment is numbered. If each pipe segment leaks, the leakage amount of each pipe segment is obtained by the difference between the size of the signal measured by the sensors of adjacent monitoring points and the set value, i.e., the imbalance of the pipe. If the i monitoring point is on the j pipe segment, the Lij of the topological matrix is 1, otherwise, the Lij is 0. The following matrices are obtained: a pipe segment length matrix l, a leakage amount matrix Q, a pipe segment weight matrix W, a monitoring point coordinate matrix J, and an associated matrix L of monitoring points and pipe segments.

4. The method according to claim 1, wherein, In calculating the maximum coverage rate, a random traversal method is used. According to the size of the number of monitoring points and the monitoring points provided after optimization of the shortest interval, a part of combinations is randomly selected. The specific information in each combination is the number of different monitoring points, and the coverage range of each pipe segment under the combination is calculated. The coverage rate is calculated by traversing each pipe segment. When a pipe segment is monitored by two monitoring points, if the length of the pipe segment is greater than c, the monitored pipe segment length is c. If the length of the pipe segment is less than c, the monitored pipe segment length is the total length of the pipe segment. When a pipe segment is monitored by one monitoring point, if the length of the pipe segment is greater than c / 2, the monitored pipe segment length is c / 2. If the length of the pipe segment is less than c / 2, the monitored pipe segment length is the total length of the pipe segment. The length of all pipes is calculated as l2, the number of monitoring points is i, and 100×i iterations are performed. The length of the pipe segment covered by the selected monitoring point is calculated as l1. The ratio l1 / l2 is taken as the size of the pipe network coverage rate. The maximum coverage rate under each number of monitoring points is calculated by comparison. Each calculation result is compared with the maximum value, and the maximum coverage rate under the number of monitoring points is finally output.

Images