Pipe network leakage prediction method, medium and terminal

Abstract

The invention is suitable for the technical field of pipe network leakage prediction, and relates to a pipe network leakage prediction method, a medium and a terminal.The method comprises the following steps that S10, influence factors influencing pipe network leakage are looked up; s20, introducing a fuzzy evaluation mechanism, and obtaining important leakage influence factors which historically influence the leakage of the pipe network in the region according to the membership degree through an expert experience library; s30, establishing a BP neural network prediction model based on PSO according to the important leakage influence factors; and S40, according to the PSO-based neural network prediction model, predicting and obtaining the safe use time of the leakage. The method is simple in process and convenient to operate, effectively improves the accuracy and real-time performance of monitoring and predicting the leakage of the pipe network, and has better benefits.

Description

Technical Field

[0001] The present invention belongs to the technical field of pipeline network leakage prediction, and in particular relates to a pipeline network leakage prediction method, medium and terminal. Background Art

[0002] The existing technology for predicting pipeline leakage generally adopts the neural network BP prediction model and the PSO particle swarm prediction model. The fastest gradient descent method in the nonlinear programming of the BP neural network prediction model is to modify the weight according to the gradient direction of the error function, and the PSO+BP neural network prediction model is to solve the minimum problem of the error function. The above methods have problems such as low learning efficiency, slow convergence speed, and easy to fall into local minimum state; a "particle" (Particle) searched in the PSO particle swarm space is a solution to the optimization problem. The optimal solution also has limitations and poor convergence, resulting in low accuracy in pipeline leakage prediction.

[0003] Patent application publication number CN116624786A provides a pipeline leakage prediction method and storage medium. This method preprocesses basic pipeline network data to obtain additional network data. This data and other network data are then input into a pre-trained nested BP neural network model, which then outputs predicted network leakage. This patent application uses a BP neural network model to predict pipeline leakage, but the prediction accuracy is low, sharing the same drawbacks as existing technologies.

[0004] Therefore, how to improve the accuracy of pipeline leakage prediction is an urgent problem to be solved by people in this technical field. Summary of the Invention

[0005] In view of the shortcomings of the existing technology, the purpose of the present invention is to provide a pipeline leakage prediction method to solve the problem of low accuracy of pipeline leakage prediction in the existing technology; in addition, the present invention also provides a pipeline leakage prediction medium and terminal.

[0006] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0007] In a first aspect, the present invention provides a method for predicting pipe network leakage, comprising the following steps:

[0008] S10. Check the factors affecting the leakage of pipe network;

[0009] S20. Introduce a fuzzy evaluation mechanism to obtain the important leakage influencing factors that have historically affected the leakage of the regional pipeline network based on the membership degree through the expert experience database;

[0010] S30, establishing a PSO-based BP neural network prediction model according to the important leakage influencing factors;

[0011] S40: Predict and obtain the safe use time of the leakage according to the PSO-based neural network prediction model.

[0012] Furthermore, the influencing factors in step S10 include user access quantity, pipe section length, pipe section diameter, pipe section material, pipe age, ambient temperature, ambient pH, soil settlement, pipe network operating pressure, road traffic volume, and pipe section pre-buried depth.

[0013] Furthermore, in step S20, the specific steps of introducing the fuzzy evaluation mechanism are as follows:

[0014] S201. Establish factor values and list all factors that affect the evaluation object;

[0015] S202, establishing a comment set, and defining the impact degree of each influencing factor as high, medium, or low;

[0016] S203, constructing a membership matrix, and determining the membership of each factor to each comment through expert system scoring or data statistics;

[0017] S204. Determine the weight of each influencing factor based on the historical experience database and the fuzzy analytic hierarchy process;

[0018] S205. Determine the order of the impact of leakage through comprehensive analysis of the degree of membership and weight, and select the top five influencing factors as important leakage influencing factors.

[0019] Furthermore, in step S30, the specific steps of establishing the PSO-based BP neural network prediction model are as follows:

[0020] S301, normalizing the five important leakage influencing factors obtained based on the fuzzy evaluation;

[0021] S302, initialize the position and velocity of the particle, set the maximum velocity and minimum velocity of the particle, set the inertia weight w, individual extreme value pbest and global extreme value gbest of the particle;

[0022] S303, mapping the particle position to the solution space and calculating the fitness value of each particle;

[0023] S304, updating the velocity and position of the particle according to the formula of the PSO algorithm;

[0024] S305: If the preset number of iterations is reached or the fitness value meets the requirement, the iteration is stopped; otherwise, the process returns to step S303;

[0025] S306: The searched optimal solution is used as the initial weight and bias, and the BP neural network is used for training. During the training process, the weight and bias of the network are adjusted until the preset number of iterations is reached or the accuracy requirement is met.

[0026] Furthermore, the important leakage influencing factors are the pipe network operating pressure, pipe age, pipe section pre-buried depth, pipe section material, and pipe section diameter.

[0027] In a second aspect, the present invention further provides a computer-readable storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the method described above is implemented.

[0028] In a third aspect, the present invention further provides an electronic terminal, comprising: a processor and a memory; the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory, so that the terminal executes the method described above.

[0029] Compared with the prior art, the pipeline network leakage prediction method, medium and terminal provided by the present invention have at least the following beneficial effects:

[0030] Existing pipeline leakage prediction technologies have limitations and poor convergence, resulting in low accuracy. The present invention has a simple process and convenient operation. By identifying factors that influence pipeline leakage, based on a fuzzy evaluation mechanism and introducing an expert database, the present invention obtains important factors affecting the pipeline network based on historical experience. These important factors are then incorporated into the PSO-based neural network prediction to infer the safe use time of the pipeline network, thereby improving the accuracy and real-time performance of pipeline leakage monitoring and prediction, and achieving good returns. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the scheme of the present invention, a brief introduction is given below to the figures required for use in the description of the embodiments. Obviously, the figures described below are some embodiments of the present invention. For ordinary technicians in this field, other figures can be obtained based on these figures without paying any creative work.

[0032] Figure 1 A flow chart of a pipeline network leakage prediction method provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0033] To facilitate understanding of the present invention, the present invention will be described more fully below with reference to the accompanying drawings. Preferred embodiments of the present invention are shown in the accompanying drawings. However, the present invention may be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a more thorough and comprehensive understanding of the present disclosure.

[0034] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0035] The present invention provides a pipeline network leakage prediction method, which is applied to the pipeline leakage prediction process. The pipeline network leakage prediction method includes:

[0036] S10. Check the factors affecting pipeline leakage; S20. Introduce a fuzzy evaluation mechanism and obtain the important leakage influencing factors that have historically affected pipeline leakage in the area based on the membership degree through the expert experience database; S30. Establish a PSO-based BP neural network prediction model based on the important leakage influencing factors; S40. Predict the safe use time of leakage based on the PSO-based neural network prediction model.

[0037] The present invention has a simple process and is easy to operate, effectively improves the accuracy and real-time performance of pipeline network leakage monitoring and prediction, and has good benefits.

[0038] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0039] The present invention provides a pipeline network leakage prediction method, which is applied to the pipeline leakage prediction process, such as Figure 1 As shown, in this embodiment, the pipeline network leakage prediction method includes:

[0040] S10. Check the factors that affect pipeline leakage.

[0041] Specifically, in this embodiment, based on the water leakage investigation report of the water supply enterprise, the causes of water leakage are recorded including user access volume, pipe section length, pipe section diameter, pipe section material, pipe age, ambient temperature, ambient pH, soil settlement, pipe network operating pressure, road traffic volume, pipe section pre-buried depth, etc.

[0042] S20. Introduce a fuzzy evaluation mechanism and obtain the important leakage influencing factors that have historically affected the pipeline leakage in the area based on the membership degree through the expert experience database.

[0043] Specifically, in this embodiment, the specific steps of step S20 are as follows:

[0044] S201. Establish factor values and list all factors that affect the evaluation object. Factors affecting pipeline network leakage include user access volume, pipe section length, pipe section diameter, pipe section material, pipe age, ambient temperature, ambient pH, soil settlement, pipeline network operating pressure, road traffic volume, pipe section pre-buried depth, etc.

[0045] In this embodiment, the first-level evaluation factor U is set to a finite value U={U1,U2,........Un}Ui∩Uj=φ(i≠j), where U represents a set of multiple factors for comprehensive evaluation of leakage factors, called a factor set; specifically,

[0046] S202. Establish a comment set V = {V1, V2..., Vm}, where V is a set of multiple judgments, called comments. The comments of the influence degree of each influencing factor are defined as high, medium, and low, V = {high, medium, low}.

[0047] S203. Construct a membership matrix. Use expert system scoring or data statistics to determine the membership of each factor to each comment. The expert system scoring is mainly based on the leakage factors of the region over the years, R = {R1, R2..., Rm}.

[0048] Specifically, according to expert scoring, the membership of user access volume is R1 = [0.20 0.30 0.50], the membership of pipe length is R2 = [0.25 0.35 0.4], the membership of pipe diameter is R3 = [0.50 0.30 0.2], the membership of pipe material is R4 = [0.60 0.20 0.20], the membership of pipe age is R5 = [0.70 0.20 0.10], the membership of ambient temperature is R6 = [0.20 0.30 0.50], the membership of ambient pH is R7 = [0.28 0.32 0.40], the membership of soil settlement is R8 = [0.32 0.37 0.31], and the membership of road traffic volume is R9 = [0.45 0.35 0.20], the membership of the embedded depth of the pipe section is R10 = [0.50 0.35 0.15], and the membership of the pipe network operating pressure is R11 = [0.80 0.15 0.05]. Therefore, the membership matrix is as follows:

[0049]

[0050] S204 , determining a weight vector, where the weight vector is A={A1, A2, A3...An}, and determining the weight of each factor based on a historical experience database and fuzzy analytic hierarchy process.

[0051] Specifically, in this embodiment, the weight of user access volume is 0.02, the weight of pipe length is 0.03, the weight of pipe diameter is 0.1, the weight of pipe material is 0.1, the weight of pipe age is 0.2, the weight of ambient temperature is 0.02, the weight of ambient pH is 0.03, the weight of soil settlement is 0.05, the weight of pipe pre-buried depth is 0.15, and the weight of pipe network operating pressure is 0.25. Therefore, the weight vector is A = (0.02 0.03 0.10 0.10 0.20 0.02 0.03 0.05 0.05 0.15 0.25).

[0052] Furthermore, in this embodiment, the weight vector A and the membership matrix R are fuzzy synthesized to obtain a comprehensive evaluation vector B:

[0053]

[0054] S205. From the comprehensive analysis of the membership matrix and weights, the impact ranking of leakage can be determined, and the five most influential factors are selected, namely pipeline network operating pressure, pipe age, pipe section pre-buried depth, pipe section material, and pipe section diameter. The five factors affecting leakage are used as input for the subsequent establishment of a prediction model.

[0055] S30. Establish a BP neural network prediction model based on PSO according to important leakage influencing factors.

[0056] Specifically, in this embodiment, the specific steps of step S30 are as follows:

[0057] S301. Perform normalization on the five important influencing factors obtained based on fuzzy evaluation, namely, pipe network operating pressure, pipe age, pipe segment pre-buried depth, pipe segment material, and pipe segment diameter. Assume that the data set is X, where each row represents a sample and each column represents an influencing factor:

[0058]

[0059] S302: Initialize the position and velocity of the particle, set the maximum velocity and minimum velocity of the particle, set the particle's inertia weight w, individual extreme value pbest, and global extreme value gbest.

[0060] S303: Map the particle positions to the solution space and calculate the fitness value of each particle. The fitness value can represent the training effect or prediction accuracy of the network.

[0061]

[0062] Where n is the number of training samples, is the ideal output value of the i-th sample, and yi is the actual output value of the i-th sample. Therefore, the smaller the training error, the lower the fitness of the particle. The particle with the lowest fitness when the algorithm iteration stops is the optimal solution to the training problem.

[0063] S304. Update the velocity and position of the particle according to the formula of the PSO algorithm:

[0064] vid(t+1)=w·vid(t)+c1·r1·(pbestid-xid(t))+c2·r2·(gbestd-xid(t));

[0065] xid(t+1)=xid(t)+vid(t+1);

[0066] Where vid(t) and xid(t) represent the velocity and position of the i-th particle in the d-th dimension, w is the inertia weight, c1 and c2 are learning factors, r1 and r2 are random numbers between [0,1], pbestid and gbestd are the individual extreme value and global extreme value of the particle, respectively.

[0067] S305: If the preset number of iterations is reached or the fitness value meets the requirement, the iteration is stopped, otherwise it returns to step S303;

[0068] S306: The searched optimal solution is used as the initial weight and bias, and the BP neural network is used for training. During the training process, the weight and bias of the network are adjusted until the preset number of iterations is reached or the accuracy requirement is met.

[0069] S40, predicting and obtaining the safe use time of the leakage according to the neural network prediction model based on PSO.

[0070] Specifically, the trained BP neural network is used for prediction. New samples are input and the predicted value of the network is output. The prediction accuracy of the model is verified by comparing the predicted value with the actual value, and the mean square error (MSE) between the predicted value and the actual value is calculated:

[0071]

[0072] Where yi is the actual value, y^i is the predicted value, and N is the number of test data samples.

[0073] In this embodiment, the fuzzy evaluation mechanism is used to select the important factors affecting the leakage of the pipe network in the region. The operating pressure of the pipe network, pipe age, pipe segment pre-buried depth, pipe segment material, and pipe diameter are obtained from the expert database mentioned above.

[0074] Collect historical data on the five important influencing factors mentioned above, as well as the corresponding safe use time of pipelines. For example, the water supply network data of a certain city is as follows:

[0075] Pipe diameter: DN100, DN200, DN300, DN400, DN500, DN600, etc.;

[0076] Pipe age: 10 years, 20 years, 30 years, etc.;

[0077] Pipe material: cast iron pipe, steel pipe, PVC pipe, etc.

[0078] Pressure during pipeline operation: 10kPa, 20kPa, 30kPa, etc.;

[0079] Pipeline pre-buried depth: 0.5m, 1m, 1.5m, 2m, 2.2m, etc.;

[0080] Safe use time: the actual recorded pipeline use time, in days;

[0081] The collected data is normalized to facilitate input into the neural network. The normalization formula is as follows:

[0082] x′=max(x)-min(x) / x-min(x);

[0083] Among them, x is the original data and x′ is the normalized data.

[0084] Construct a PSO-BP neural network model: initialize the weights and biases of the neural network: the number of input layer nodes is 5 (corresponding to 5 influencing factors), the number of hidden layer nodes is 5, the number of output layer nodes is 1 (predicting safety time), and the weights and biases are initialized to random values.

[0085] Set the parameters of the PSO algorithm: number of particles: 20, maximum number of iterations: 100, learning factors: c1 = 2, c2 = 2, maximum speed: 1.0, minimum speed: -1.0, maximum boundary: 2.0, minimum boundary: -2.0.

[0086] Initialize the particle position and velocity, calculate the fitness of each particle (i.e., the prediction error of the neural network), update the individual optimal position and the global optimal position, and update the particle's velocity and position.

[0087] BP neural network training: Use the training data set to train the neural network, adjust the weights and biases through the back propagation algorithm to minimize the error between the predicted value and the actual value, and repeat the PSO and BP training process until the maximum number of iterations is reached or the stopping condition is met.

[0088] Model Validation: Use the test dataset to verify the model's predictive performance. Calculate the mean squared error (MSE) between the predicted values and the actual values:

[0089]

[0090] Where yi is the actual value, y^i is the predicted value, and N is the number of test data samples.

[0091] Taking the pipelines in a certain city as an example, the prediction results of the pipeline network leakage prediction method provided by the embodiment of the present invention are compared with the actual results:

[0092] The specific data of a city’s pipelines are as follows:

[0093] Pipe diameter: DN600; Pipe age: 17 years; Pipe material: cast iron pipe; Pipe operating pressure: 19.6 kPa; Pipe buried depth: 1.3 meters; Actual safe use time: 6043 days;

[0094] After normalizing the above data, the trained PSO-BP neural network model was input to predict the safe use time: 6057.66 days, which is very close to the actual safe use time of 6043 days, proving the feasibility of the model, with MSE = 0.997.

[0095] Specifically, in this embodiment, MATLAB software is used for model training, and some specific codes are as follows:

[0096]

[0097]

[0098]

[0099]

[0100]

[0101]

[0102]

[0103]

[0104]

[0105] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, any one of the methods in the embodiment is implemented.

[0106] An embodiment of the present invention further provides an electronic terminal, comprising: a processor and a memory; the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory, so that the terminal executes any one of the methods in this embodiment.

[0107] Regarding the computer-readable storage medium in this embodiment, those skilled in the art will appreciate that all or part of the steps in the aforementioned method embodiments can be implemented using hardware associated with the computer program. The aforementioned computer program can be stored in a computer-readable storage medium. When executed, the program performs the steps in the aforementioned method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0108] The electronic terminal provided in this embodiment includes a processor, a memory, a transceiver and a communication interface. The memory and the communication interface are connected to the processor and the transceiver and complete communication with each other. The memory is used to store computer programs, the communication interface is used for communication, and the processor and the transceiver are used to run computer programs so that the electronic terminal executes the various steps of the above method.

[0109] The pipeline leakage prediction method described in the above embodiment is compared with the existing technology. The existing pipeline leakage prediction technology has limitations and poor convergence, resulting in low pipeline leakage prediction accuracy. The present invention has a simple process and convenient operation. It identifies the factors affecting pipeline leakage, introduces an expert database based on a fuzzy evaluation mechanism, and obtains important factors affecting the pipeline based on historical experience. These important factors are added to the PSO-based neural network prediction to infer the safe use time of the pipeline network, thereby improving the accuracy and real-time performance of pipeline leakage monitoring and prediction, and has better benefits.

[0110] Obviously, the embodiments described above are only preferred embodiments of the present invention, rather than all embodiments. The preferred embodiments of the present invention are given in the accompanying drawings, but they do not limit the patent scope of the present invention. The present invention can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present invention more thorough and comprehensive. Although the present invention has been described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions described in the aforementioned specific embodiments, or to make equivalent replacements for some of the technical features therein. Any equivalent structure made using the contents of the present invention specification and drawings, directly or indirectly used in other related technical fields, is also within the scope of patent protection of the present invention.

Claims

1. A method for predicting pipe network leakage, characterized in that: The following steps are involved: S10. Check the factors affecting the leakage of pipe network; S20. Introduce a fuzzy evaluation mechanism to obtain the important leakage influencing factors that have historically affected the leakage of the regional pipeline network based on the membership degree through the expert experience database; S30, establishing a PSO-based BP neural network prediction model according to the important leakage influencing factors; S40: Predict and obtain the safe use time of the leakage according to the PSO-based neural network prediction model.

2. A pipeline network leakage prediction method according to claim 1, characterized in that: The influencing factors in step S10 include user access quantity, pipe section length, pipe section diameter, pipe section material, pipe age, ambient temperature, ambient pH, soil settlement, pipe network operating pressure, road traffic volume, and pipe section pre-buried depth.

3. A pipeline network leakage prediction method according to claim 2, characterized in that: In step S20, the specific steps of introducing the fuzzy evaluation mechanism are as follows: S201. Establish factor values and list all factors that affect the evaluation object; S202, establishing a comment set, and defining the impact degree of each influencing factor as high, medium, or low; S203, constructing a membership matrix, and determining the membership of each factor to each comment through expert system scoring or data statistics; S204. Determine the weight of each influencing factor based on the historical experience database and the fuzzy analytic hierarchy process; S205. Determine the order of the impact of leakage through comprehensive analysis of the degree of membership and weight, and select the top five influencing factors as important leakage influencing factors.

4. A pipeline network leakage prediction method according to claim 3, characterized in that: In step S30, the specific steps of establishing the PSO-based BP neural network prediction model are as follows: S301, normalizing the five important leakage influencing factors obtained based on the fuzzy evaluation; S302, initialize the position and velocity of the particle, set the maximum velocity and minimum velocity of the particle, set the inertia weight w, individual extreme value pbest and global extreme value gbest of the particle; S303, mapping the particle position to the solution space and calculating the fitness value of each particle; S304, updating the velocity and position of the particle according to the formula of the PSO algorithm; S305: If the preset number of iterations is reached or the fitness value meets the requirement, the iteration is stopped; otherwise, the process returns to step S303; S306: The searched optimal solution is used as the initial weight and bias, and the BP neural network is used for training. During the training process, the weight and bias of the network are adjusted until the preset number of iterations is reached or the accuracy requirement is met.

5. A pipeline network leakage prediction method according to claim 4, characterized in that: The important factors affecting leakage are pipeline network operating pressure, pipe age, pipe segment pre-buried depth, pipe segment material, and pipe segment diameter.

6. A computer-readable storage medium, characterized in that The storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

7. An electronic terminal, characterized in that: include: processor and memory; The memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory, so that the terminal performs the method according to any one of claims 1 to 5.

Images