Multi-source data synchronous coupling analysis-based stagnant gas explosion risk grading and classification dynamic evaluation early warning model establishment method

Through the synchronous coupling analysis of multi-source data and fuzzy comprehensive evaluation method, a risk assessment model for gas stagnant explosion pipe was established, which solved the problem of difficult prediction of gas stagnant explosion pipe in long-distance pressure water transmission pipeline systems, and achieved changes in pipeline material strength Accurate assessment and future risk prediction improve engineering safety.

CN118886702APending Publication Date: 2024-11-01ZHEJIANG UNIV OF WATER RESOURCES & ELECTRIC POWER
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202410898521.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-05
Publication Date
2024-11-01

AI Technical Summary

Technical Problem

In long-distance pressure water supply pipeline systems, the risk of stagnant gas-burst pipes is difficult to predict in real time, resulting in safety risks. The prior art cannot effectively identify and evaluate the impact of changes in pipeline material strength on the risk of burst pipes.

Method used

The dynamic evaluation and warning model of risk classification of stagnant gas explosion tube based on synchronous coupling analysis of multi-source data is adopted, combined with heavy pressure data, temperature stress data and corrosion degree data, and the impact assessment model is established through the intensity prediction network and fuzzy comprehensive evaluation method, and the impact assessment model is established and determined. Current and future explosion risk levels.

Benefits of technology

It improves the prediction accuracy of the risk of stagnant gas explosion pipes, can more comprehensively evaluate the pipeline risk level, provide real-time early warning, and enhances engineering safety.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118886702A_ABST
    Figure CN118886702A_ABST
Patent Text Reader

Abstract

The invention discloses a multi-source data synchronous coupling analysis-based stagnant gas pipe explosion risk grading and classification dynamic evaluation early warning model establishment method, and relates to the technical field of pipe explosion analysis, and the method comprises the following steps: obtaining heavy pressure data, temperature stress and corrosion degree according to a preset time interval; combining the corresponding change of each time period to obtain the pipeline material strength of the corresponding time period; inputting the pipeline material strength in the historical time period into the strength prediction network to obtain the material residual strength in the current time period; the method comprises the following steps: establishing an influence evaluation model for influence data influencing a pipe explosion risk level, performing data acquisition on the influence data, performing evaluation in the influence evaluation model through a fuzzy comprehensive evaluation method in combination with material residual strength to obtain a current risk level, establishing a judgment matrix, and predicting a future pipe explosion risk level according to the current risk level. According to the method, the material strength change caused by the fact that the pipeline is influenced by other factors for a long time is considered, comprehensive evaluation is carried out in combination with a fuzzy comprehensive evaluation method, and the gas explosion risk prediction is more accurate.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of pipe burst analysis, and more specifically to a method for establishing a risk classification and dynamic assessment early warning model for gas stagnation pipe burst based on synchronous coupling analysis of multi-source data. Background Art

[0002] Water is the source of life and a limited resource that is indispensable to social development and human life. It is closely related to human reproduction and survival. As a major water delivery method, pressurized pipeline water delivery has the characteristics of small evaporation and leakage losses along the way, low risk of water pollution, and less interference along the project line. It is conducive to ensuring water quantity and quality, and has a wide range of applications. In long-distance pressurized water delivery projects, multi-undulating water delivery systems account for a large proportion, which are characterized by long pipelines and large terrain undulations.

[0003] However, long-distance pressurized water pipeline systems have the characteristics of long water pipelines, undulating terrain, and complex water flow conditions. Studies have shown that the water in long-distance pressurized water pipeline systems naturally contains gas, which may be retained in the pipeline and accumulate into trapped gas masses. During the hydraulic transition process, it is very easy to induce dangerous gas-containing water hammer, which may lead to pipe bursts. Therefore, it is very necessary to conduct research on the mechanism of pipe bursts caused by gas stagnation, identify the risk of pipe bursts caused by gas stagnation, formulate protective measures, and ensure project safety.

[0004] Therefore, how to predict the risk of gas stagnation and pipe burst in real time is an urgent problem to be solved by those skilled in the art. Summary of the invention

[0005] In view of this, the present invention provides a method for establishing a dynamic assessment and early warning model for grading and classification of risk of gas stagnation and pipe burst based on synchronous coupling analysis of multi-source data. The method takes into account the changes in material strength of the pipeline caused by long-term influence of other factors, and combines the fuzzy comprehensive evaluation method to comprehensively evaluate the fuzzy relationship matrix and weight vector to determine the current risk level vector, thereby making the prediction of gas stagnation and pipe burst risk more accurate.

[0006] In order to achieve the above object, the present invention adopts the following technical solution:

[0007] A method for establishing a risk classification and dynamic assessment and early warning model for gas stagnation and pipe burst based on multi-source data synchronous coupling analysis, comprising:

[0008] Obtaining heavy pressure data, temperature stress data, and corrosion degree data according to preset time intervals; combining the pressure change of the heavy pressure data, the stress change of the temperature stress data, and the degree change of the corrosion degree data in each period to obtain the pipeline material strength of the corresponding period; inputting the pipeline material strength of multiple historical periods into the strength prediction network to obtain the material residual strength of the current period;

[0009] An impact assessment model is established for the impact data affecting the pipe burst risk level, data collection is performed on the impact data of the pipe burst risk level, and the collected data is evaluated in the impact assessment model by using a fuzzy comprehensive evaluation method in combination with the residual strength of the material to obtain the current pipe burst risk level;

[0010] A judgment matrix is ​​established based on the Markov model, and the future pipe burst risk level is predicted based on the current pipe burst risk level.

[0011] Preferably, it also includes: using a correction coefficient to correct the material residual strength of the current time period to obtain the actual material residual strength; the correction coefficient is obtained by the ratio between the known material residual strength and the pipeline material strength actually calculated in the corresponding time period.

[0012] Preferably, the strength prediction network adopts a mean square error loss function, and the mean square error loss is a loss mean calculated by weighted summation of the mean square error losses between the input and output data of each training sample; the weight of each training sample is positively correlated with the material strength evaluation coefficient of the pipeline in the corresponding time period, and the material strength evaluation coefficient of the pipeline material strength corresponding to each time period is assigned a weight to its mean square error loss, all material strength evaluation coefficients constitute an evaluation sequence, and the evaluation sequence is normalized, and the coefficient of the normalized evaluation sequence is used as the weight of the corresponding training sample, and the mean square error loss function is:

[0013]

[0014] in, is the i-th output data; is the i-th input data; Q i is the weight corresponding to the input data, and W is the total number of input data.

[0015] Preferably, the material strength evaluation coefficient specifically includes: forming a strength sequence of the pipeline material strengths corresponding to multiple historical time periods, respectively calculating the difference between the pipeline material strength in the current time period and each pipeline material strength in the strength sequence, obtaining the sum of the differences, and obtaining the material strength evaluation coefficient of the pipeline material strength according to the sum of the differences.

[0016] Preferably, the training set of the strength prediction network includes M training samples, and each training sample includes a sequence consisting of the pipeline material strengths of the pipeline corresponding to N time periods, wherein one time period corresponds to one pipeline material strength.

[0017] Preferably, obtaining the current pipe burst risk level specifically includes:

[0018] An impact assessment model is constructed based on the type and location of the impact data affecting the pipe burst risk level and the residual strength of the material to determine the impact degree of the impact data affecting the pipe burst risk level;

[0019] By collecting data affecting the risk level of pipe burst, the collected data is processed by fuzzy comprehensive evaluation method according to the impact assessment model to obtain a current risk level vector;

[0020] The current risk level of the pipeline burst is determined by the current risk level vector.

[0021] Preferably, the determining of the influence degree of the influencing data affecting the pipe burst risk level specifically includes:

[0022] Establish a risk assessment structure model for the types of influencing data and monitoring locations that affect pipe burst risk levels;

[0023] Comparing the residual strength of the material at the monitoring location and the degree of influence between various influencing data on the basis of the risk assessment structural model to form a judgment matrix;

[0024] A consistency check is performed on the judgment matrix, and a weight vector of the material residual strength and a weight vector of each influencing data are obtained after normalization processing is performed on the judgment matrix that passes the consistency check.

[0025] Preferably, obtaining the current risk level vector specifically includes: collecting data affecting the risk of pipe burst through sensors, performing membership vector quantization processing on the collected data and the residual strength of the material to obtain a fuzzy relationship matrix corresponding to the data in the current time period, performing fuzzy comprehensive evaluation through the fuzzy relationship matrix and the weight vector, and determining the current risk level vector.

[0026] Preferably, the influencing data specifically include: heavy pressure data, temperature stress data, corrosion degree data, water hammer pressure data, and construction technology data.

[0027] It can be seen from the above technical solution that, compared with the prior art, the present invention discloses a method for establishing a dynamic assessment and early warning model for stagnation and bursting of gas pipes based on synchronous coupling analysis of multi-source data. It combines multiple data sources such as heavy pressure data, temperature stress data, and corrosion degree data. By synchronously coupling and analyzing these data, the risk level of the pipeline can be more comprehensively evaluated; by establishing a strength prediction network, the strength of the pipeline material in the current period can be predicted based on historical data, thereby obtaining the residual strength of the material in the current period; the residual strength of the material is corrected by a correction coefficient, thereby improving the accuracy of the prediction; an impact assessment model is established, and the current risk level of the pipe burst can be obtained by collecting data on the impact data of the pipe burst risk level and evaluating it in combination with the residual strength of the material; by establishing a judgment matrix, the future risk level of the pipe burst can also be predicted.

[0028] By forming a strength sequence of pipeline material strengths in historical periods and calculating the difference between the material strength in the current period and each strength in the sequence, the material strength evaluation coefficient can be obtained. This evaluation coefficient can more accurately reflect the material strength of the current period; by quantizing the influencing data into membership vectors and combining them with weight vectors for fuzzy comprehensive evaluation, the current risk level vector can be obtained. This evaluation method can more comprehensively consider various influencing factors and improve the accuracy of the evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] 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 embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0030] Figure 1 A diagram of the method steps provided by the present invention;

[0031] Figure 2 This is a pipeline risk assessment structure diagram provided by the present invention. DETAILED DESCRIPTION

[0032] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0033] The embodiment of the present invention discloses a method for establishing a dynamic assessment and early warning model for stagnation and pipe burst risk classification based on multi-source data synchronous coupling analysis, such as Figure 1 As shown, including:

[0034] Obtaining heavy pressure data, temperature stress data, and corrosion degree data according to preset time intervals; combining the pressure change of heavy pressure data, the stress change of temperature stress data, and the degree change of corrosion degree data in each period to obtain the pipeline material strength of the corresponding period; inputting the pipeline material strength of multiple historical periods into the strength prediction network to obtain the material residual strength of the current period;

[0035] An impact assessment model is established for the impact data affecting the pipe burst risk level, and data collection is performed on the impact data of the pipe burst risk level. The collected data is combined with the residual strength of the material and evaluated in the impact assessment model through the fuzzy comprehensive evaluation method to obtain the current pipe burst risk level;

[0036] An evaluation matrix is ​​established based on the Markov model, and the future pipe burst risk level is predicted based on the current pipe burst risk level.

[0037] Among them, heavy pressure data refers to the high pressure or high load that the pipeline bears for a long time, which is caused by the high pressure of the medium transported by the pipeline or the environmental conditions in which the pipeline is located.

[0038] In a specific embodiment, it also includes: using a correction coefficient to correct the material residual strength of the current period to obtain the actual material residual strength; the correction coefficient is obtained by the ratio between the known material residual strength and the pipeline material strength actually calculated in the corresponding period.

[0039] Specifically, the strength prediction network adopts the TCN network. The training process of the TCN network includes: firstly, obtaining a training set, which includes multiple training samples. The data included in each training sample is a sequence composed of the pipeline material strengths of the pipeline corresponding to N time periods. One time period corresponds to one pipeline material strength. Then, taking a training sample as an example, let the pipeline material strength of the pipeline in the first time period be r 1 , the pipeline material strength of the pipeline in the Nth period is r N , the pipeline material strength of each period from the first period to the Nth period is formed into a stable sequence {r1, r2, r3, ...r N}, remove the extreme values ​​in the stable sequence to obtain a new evaluation sequence, and then use the new evaluation sequence as a training sample; input the new evaluation sequence in each training sample into the TCN network as a feature sequence, and use the next value of the feature sequence as a label so that the TCN network can learn the next predicted value under the current sequence mode, and then obtain the remaining feature sequences, and repeat the training; the loss function of the TCN network adopts the mean square error loss function. Using the stable sequence obtained in real time to train the TCN network enables the TCN network to predict the remaining strength of the material of the pipeline in the current period.

[0040] The trained TCN network can be used to obtain the remaining material strength of the pipeline in the next period. Since there is a difference between the predicted value and the actual calculated pipeline material strength, the ratio between the known remaining material strength and the actual distributed pipeline material strength in the corresponding period is used as the error correction coefficient.

[0041] In a specific embodiment, the strength prediction network adopts a mean square error loss function, and the mean square error loss is the loss mean calculated by weighted summation of the mean square error loss between the input and output data of each training sample; the weight of each training sample is positively correlated with the material strength evaluation coefficient of the pipeline in the corresponding period, and the material strength evaluation coefficient of the pipeline material strength corresponding to each period is assigned a weight to its mean square error loss, and all material strength evaluation coefficients constitute an evaluation sequence, and the evaluation sequence is normalized, and the coefficient of the normalized evaluation sequence is used as the weight of the corresponding training sample, and the mean square error loss function is:

[0042]

[0043] in, is the i-th output data; is the i-th input data; Q i is the weight corresponding to the input data, and W is the total number of input data. The material strength evaluation coefficient corresponding to the pipeline material strength in each period is used as the weight, the purpose of which is to ensure the convergence of the loss function, reduce the loss through continuous training, and improve the accuracy of the prediction trend.

[0044] In a specific embodiment, the material strength evaluation coefficient specifically includes: forming a strength sequence from the pipeline material strengths corresponding to multiple historical time periods, respectively calculating the difference between the pipeline material strength in the current time period and each pipeline material strength in the strength sequence, obtaining the sum of the differences, and obtaining the material strength evaluation coefficient of the pipeline material strength according to the sum of the differences.

[0045] The evaluation sequence R = {r 2 ,...,r i}, respectively calculate the material strength evaluation coefficient corresponding to each pipeline material strength. Specifically, respectively calculate the difference D (r i ,R), get the sum of the differences, and obtain the material strength evaluation coefficient corresponding to the current pipeline material strength according to the sum of the differences, that is, Among them, ∑D(r i ,R) is the sum of the differences, which reflects the fluctuation of the current pipeline material strength compared with other time periods. That is, the smaller the sum of the differences, the smaller the fluctuation, and the larger the corresponding evaluation coefficient, which also indicates that the detection result of the current pipeline material strength is more reliable and accurate.

[0046] In a specific embodiment, the training set of the strength prediction network includes M training samples, and each training sample includes a sequence consisting of pipeline material strengths corresponding to pipelines in N time periods, wherein one time period corresponds to one pipeline material strength.

[0047] In a specific embodiment, obtaining the current pipe burst risk level specifically includes:

[0048] An impact assessment model is constructed based on the type and location of the impact data affecting the pipe burst risk level and the residual strength of the material to determine the impact degree of the impact data affecting the pipe burst risk level;

[0049] By collecting data on the impact of pipe burst risk level, the current risk level vector is obtained by processing the collected data through fuzzy comprehensive evaluation method according to the impact assessment model;

[0050] The current risk level of the pipeline burst is determined by the current risk level vector.

[0051] In a specific embodiment, a two-level fuzzy comprehensive evaluation method is used, the second level is the impact of the same influencing data on the pipe burst risk level at different observation points. The first level evaluates different influencing data that affect the pipeline status based on the second level.

[0052] In a specific embodiment, determining the influence degree of the influencing data affecting the pipe burst risk level specifically includes:

[0053] Establish a risk assessment structure model for the types of influencing data and monitoring locations that affect pipe burst risk levels;

[0054] On the basis of the risk assessment structural model, the residual strength of the material at the monitoring location and the degree of influence between various influencing data are compared to form a judgment matrix;

[0055] The judgment matrix is ​​checked for consistency, and the judgment matrix that passes the consistency check is normalized to obtain the weight vector of the material residual strength and the weight vector of each influencing data.

[0056] In a specific embodiment, the analytic hierarchy process (AHP) is applied to determine the weights of each influencing data and the residual strength of the material, wherein the steps of determining each weight are:

[0057] Establish the structure of pipeline risk assessment model: set the final goal as the current burst risk level of the pipeline, the first layer is the various influencing data affecting the pipeline risk level and the residual strength of the materials at the corresponding positions, and the second layer is the impact of the same influencing data at different positions on the pipeline risk level. The assessment hierarchy established in this way is not only logically rigorous, but also conforms to the evaluation characteristics of the pipeline status, such as Figure 2 shown.

[0058] Construct a judgment matrix: Based on the pipeline risk assessment model, compare the relative importance of each layer to obtain the weight of each layer, and then construct a judgment matrix.

[0059] Construct judgment matrix A i for:

[0060]

[0061] Among them, r ij It indicates the importance of the i-th influencing factor relative to the j-th factor, where the influencing factors in factors i and j include the residual strength of the material, and the relative importance is shown in Table 1.

[0062] Table 1 Relative importance of AHP

[0063]

[0064]

[0065] In a specific embodiment, obtaining the current risk level vector specifically includes: collecting data affecting the risk of pipe burst through sensors, performing membership vector quantization processing on the collected data and the residual strength of the material, obtaining a fuzzy relationship matrix corresponding to the data in the current time period, and performing fuzzy comprehensive evaluation through the fuzzy relationship matrix and the weight vector to determine the current risk level vector.

[0066] In a specific embodiment, the membership degree of each evaluation physical quantity is determined by using fuzzy statistics, and the impact data affecting the risk of pipe burst is collected by sensors. The threshold range determined after data processing of a single evaluation impact data is divided into k intervals for evaluation respectively, and the evaluation result of each interval is expressed as a membership vector X. p Indicates. For L data measured in a certain period of time, let the number of times it belongs to each interval be n p(p=1,2,3...), recorded as the weight of the interval, the membership vector of the factor in this time period after weighted average is:

[0067]

[0068] Since a two-layer model structure was established, a two-level fuzzy comprehensive evaluation was adopted.

[0069] First-level fuzzy comprehensive evaluation: The first-level fuzzy comprehensive evaluation is to evaluate the second level. The present invention performs fuzzy comprehensive evaluation on the monitoring pipeline position, that is, each monitored position that affects the data being measured.

[0070] Establish the second-level weight set, determine the second-level fuzzy relationship matrix, and process each single impact data through membership vectorization to form the fuzzy relationship matrix E i .

[0071] Carry out the first-level fuzzy comprehensive evaluation: the second-level fuzzy relationship matrix E i and the weight vector of the second layer The weighted average model M(·,+) is used for first-level comprehensive evaluation:

[0072]

[0073] Then to B i 'Normalization processing:

[0074] Bi=[bi1,bi2,bi3,bi4,bi5];

[0075] Among them, i=0, 1, 2...n.

[0076] Secondary fuzzy comprehensive evaluation: The secondary fuzzy comprehensive evaluation is an evaluation of the first layer, which is a fuzzy comprehensive evaluation of the monitored pipeline sections, that is, the measured influencing data.

[0077] Establish the first layer weight set: Establish the first layer weight set The method is the same as the second layer;

[0078] Determine the first-layer fuzzy relationship matrix: The first-layer fuzzy relationship matrix is ​​the second-layer evaluation result B i The elements in , namely:

[0079] R=(B1,B2,B3,B4) T ;

[0080] Perform secondary fuzzy comprehensive evaluation: The first-layer fuzzy relationship matrix R and the first-layer weight vector The weighted average model M(·,+) is used for secondary comprehensive evaluation:

[0081]

[0082] In a specific embodiment, in establishing a judgment matrix, the risk i of the system in time period t is represented as X t =i, and the risk j of the system in period t+1 is expressed as X t+1 =j, using the Markov model method, the probability of judging in time period t is expressed as:

[0083] g ij (t) = G{X t =i|X t+1 =j},i,j∈S;

[0084] Where g ij (t) is the probability of the system judging risk i at t to risk j at t+1. In any period t, we aggregate the probability of the system judging risk i at t to risk j at t+1 to form the system's judgment probability matrix:

[0085]

[0086] This matrix satisfies: All probabilities are non-negative 0≤g ij ≤1(i,j=0,1…n); the sum of the probabilities of each row is 1:

[0087] Among them, the pipeline risk assessment probability matrix usually adopts the risk statistics method:

[0088]

[0089] Among them, g ij represents the probability that the pipeline transfers from risk i to risk j in a given time interval, X ij represents the number of pipelines transferred from risk i to risk j in a given time interval, X i It indicates the number of pipelines with risk i in a given time interval.

[0090] In a specific embodiment, the influencing data specifically include: heavy pressure data, temperature stress data, corrosion degree data, water hammer pressure data, and construction technology data.

[0091] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0092] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for establishing a risk classification and dynamic assessment and early warning model for gas stagnation and pipe burst based on multi-source data synchronous coupling analysis, characterized in that: include: Obtaining heavy pressure data, temperature stress data, and corrosion degree data according to preset time intervals; Combining the pressure change of the heavy pressure data, the stress change of the temperature stress data and the degree change of the corrosion degree data in each period, the pipeline material strength of the corresponding period is obtained; the pipeline material strength of multiple historical periods is input into the strength prediction network to obtain the material residual strength of the current period; An impact assessment model is established for the impact data affecting the pipe burst risk level, data collection is performed on the impact data of the pipe burst risk level, and the collected data is evaluated in the impact assessment model by using a fuzzy comprehensive evaluation method in combination with the residual strength of the material to obtain the current pipe burst risk level; A judgment matrix is ​​established based on the Markov model, and the future pipe burst risk level is predicted based on the current pipe burst risk level.

2. According to claim 1, a method for establishing a risk classification and dynamic assessment and early warning model for gas stagnation and pipe burst based on multi-source data synchronous coupling analysis is characterized in that: It also includes: using a correction coefficient to correct the material residual strength of the current period to obtain the actual material residual strength; the correction coefficient is obtained by the ratio between the known material residual strength and the pipeline material strength actually calculated in the corresponding period.

3. According to claim 1, a method for establishing a risk classification and dynamic assessment and early warning model for gas stagnation and pipe burst based on multi-source data synchronous coupling analysis is characterized in that: The strength prediction network adopts a mean square error loss function, which is a loss mean calculated by weighted summation of the mean square error losses between the input and output data of each training sample; the weight of each training sample is positively correlated with the material strength evaluation coefficient of the pipeline in the corresponding period, and the material strength evaluation coefficient of the pipeline material strength corresponding to each period is assigned a weight to its mean square error loss, all material strength evaluation coefficients constitute an evaluation sequence, the evaluation sequence is normalized, and the coefficient of the normalized evaluation sequence is used as the weight of the corresponding training sample, and the mean square error loss function is: in, is the i-th output data, is the i-th input data, Q i is the weight corresponding to the input data, and W is the total number of input data.

4. The method for establishing a risk classification and dynamic assessment and early warning model for gas stagnation and pipe burst based on multi-source data synchronous coupling analysis according to claim 3 is characterized in that: The material strength evaluation coefficient specifically includes: forming a strength sequence of the pipeline material strengths corresponding to multiple historical time periods, respectively calculating the difference between the pipeline material strength in the current time period and each pipeline material strength in the strength sequence, obtaining the sum of the differences, and obtaining the material strength evaluation coefficient of the pipeline material strength according to the sum of the differences.

5. According to claim 1, a method for establishing a risk classification and dynamic assessment and early warning model for gas stagnation and pipe burst based on multi-source data synchronous coupling analysis is characterized in that: The training set of the strength prediction network includes M training samples, and each training sample includes a sequence consisting of the pipeline material strengths of the pipeline corresponding to N time periods, wherein one time period corresponds to one pipeline material strength.

6. The method for establishing a risk classification and dynamic assessment and early warning model for gas stagnation and pipe burst based on multi-source data synchronous coupling analysis according to claim 1 is characterized in that: The obtaining of the current pipe burst risk level specifically includes: An impact assessment model is constructed based on the type and location of the impact data affecting the pipe burst risk level and the residual strength of the material to determine the impact degree of the impact data affecting the pipe burst risk level; By collecting data affecting the risk level of pipe burst, the collected data is processed by fuzzy comprehensive evaluation method according to the impact assessment model to obtain a current risk level vector; The current risk level of the pipeline burst is determined by the current risk level vector.

7. The method for establishing a risk classification and dynamic assessment and early warning model for gas stagnation and pipe burst based on multi-source data synchronous coupling analysis according to claim 6 is characterized in that: The influence degree of the influencing data affecting the pipe burst risk level determined specifically includes: Establish a risk assessment structure model for the types of influencing data and monitoring locations that affect pipe burst risk levels; Comparing the residual strength of the material at the monitoring location and the degree of influence between various influencing data on the basis of the risk assessment structural model to form a judgment matrix; A consistency check is performed on the judgment matrix, and a weight vector of the material residual strength and a weight vector of each influencing data are obtained after normalization processing is performed on the judgment matrix that passes the consistency check.

8. The method for establishing a risk classification and dynamic assessment and early warning model for gas stagnation and pipe burst based on multi-source data synchronous coupling analysis according to claim 7 is characterized in that: The obtaining of the current risk level vector specifically includes: collecting data affecting the risk of pipe burst through sensors, performing membership vector quantization processing on the collected data and the residual strength of the material, obtaining a fuzzy relationship matrix corresponding to the data in the current time period, and performing fuzzy comprehensive evaluation through the fuzzy relationship matrix and the weight vector to determine the current risk level vector.

9. The method for establishing a risk classification and dynamic assessment and early warning model for gas stagnation and pipe burst based on multi-source data synchronous coupling analysis according to claim 1 is characterized in that: The influencing data specifically include: heavy pressure data, temperature stress data, corrosion degree data, water hammer pressure data, and construction technology data.

Citation Information

Cited By

  • Gas pipeline operation risk assessment method and system

    CN120317692A

  • Power plant boiler risk assessment method and system

    CN122658042A