Multi-dimensional joint distribution-based multi-defect pipeline evaluation method, device and equipment

Abstract

The invention relates to the field of pipeline safety, and provides a multi-dimensional joint distribution-based multi-defect pipeline evaluation method, device and equipment, and the method comprises the steps: obtaining input data, inputting the input data into a constructed probability distribution model of a pipeline under the action of a single defect, and outputting the failure probability of each defect of the pipeline and the probability distribution of each defect; the constructed probability distribution model of the pipeline under the action of the single defect is constructed according to actual pipeline detection data, historical failure data and related evaluation standards of the pipeline defect; according to the probability distribution of each defect, a multi-dimensional joint probability distribution model among the defects is constructed based on the relationship between a Vine Copula function and the reliability of the series pipeline system; and based on the inter-defect multi-dimensional joint probability distribution model, determining the reliability of the pipeline according to the failure probability of the pipeline under the action of the single defect. According to the invention, the problem that the reliability evaluation result of the pipeline is not accurate enough in the prior art is solved, and the effect of accurately evaluating the reliability of the pipeline is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of pipeline safety, and in particular to a method, device and equipment for evaluating a pipeline containing multiple defects based on multi-dimensional joint distribution. Background Art

[0002] In the field of energy storage and transportation, pipeline transportation has become the most important mode of transportation due to its wide coverage and high safety. However, as the use time of oil and gas pipelines increases, their operational safety issues have gradually become prominent, especially when pipelines are subjected to the combined effects of multiple failure mechanisms, such as corrosion, wear, impact and fatigue, which lead to frequent leakage accidents.

[0003] There are currently a variety of evaluation methods and technologies for the failure problem of pipeline systems with multiple defects. Researchers usually evaluate the reliability of systems with multiple failure modes according to the "weakest link principle" or "independence assumption principle". Among them, the weakest link principle assumes that the failure modes are completely correlated, and the reliability of the system is determined by the weakest part in the system, that is, the system reliability is equal to the smallest of the reliabilities of each subsystem. The independence assumption principle assumes that the failure modes are independent of each other, and the reliability of the system is equal to the product of the reliabilities of each subsystem, which is also the traditional series reliability method.

[0004] In recent years, with the deepening of reliability research, domestic and foreign scholars have begun to pay attention to the correlation between multiple failure modes, and try to more accurately evaluate the reliability of pipelines with multiple defects by constructing multi-dimensional joint probability distribution. Although the existing technology has made certain progress in the reliability evaluation of pipelines with multiple defects, there are still obvious defects. The weakest link principle and the independence assumption principle are extreme assumption methods. The former often overestimates the reliability, while the latter underestimates the reliability. Both cannot accurately reflect the actual reliability of pipeline systems with multiple defects. Summary of the invention

[0005] The present invention provides a method, device and equipment for evaluating a pipeline containing multiple defects based on multi-dimensional joint distribution, which solves the problem that the pipeline reliability evaluation results in the prior art are not accurate enough and achieves the effect of accurately evaluating the pipeline reliability.

[0006] The present invention provides a pipeline evaluation method containing multiple defects based on multi-dimensional joint distribution, comprising the following steps:

[0007] Acquire input data, input the input data into a constructed probability distribution model of a pipeline under a single defect, and output the failure probability of each defect of the pipeline and the probability distribution of each defect; the constructed probability distribution model of the pipeline under a single defect is constructed according to actual pipeline inspection data, historical failure data and pipeline defect-related evaluation standards;

[0008] According to the probability distribution of each defect, based on the relationship between the Vine Copula function and the reliability of the series pipeline system, a multi-dimensional joint probability distribution model between defects is constructed;

[0009] Based on the multi-dimensional joint probability distribution model between defects, the reliability of the pipeline is determined according to the failure probability of the pipeline under the action of the single defect.

[0010] According to a multi-defect pipeline evaluation method based on multi-dimensional joint distribution provided by the present invention, the input data is obtained, the input data is input into a constructed probability distribution model of a pipeline under a single defect, and the failure probability of each defect of the pipeline and the probability distribution of each defect are output, which specifically includes: obtaining basic random variables that affect the result state of the pipeline; obtaining the probability distribution characteristics of the basic random variables as input data; inputting the input data into the pipeline limit state equation to obtain an output result; the pipeline limit state equation is determined based on the evaluation standard related to pipeline defects; judging whether the execution condition of the failure probability model of the pipeline under a single defect is met according to the output result of the pipeline limit state equation; if the execution condition is met, then based on the failure probability model of the pipeline under a single defect, the failure probability of the pipeline under a single defect is determined according to the input data; the failure probability of the pipeline under the effect of each defect is obtained one by one, and the probability distribution of each defect with the pipeline limit state equation as a random variable is further fitted according to the failure probability of each defect.

[0011] According to a multi-dimensional joint distribution-based pipeline evaluation method provided by the present invention, the failure probability model of the pipeline under the action of a single defect is based on the input data to determine the failure probability of the pipeline under the action of a single defect, which specifically includes: using the Monte Carlo method to generate corresponding random variables according to the probability distribution characteristics of multiple basic random variables as sample data of the solution; substituting the sample data into the limit state equation of the pipeline, based on the law of large numbers, counting the number of failures, and obtaining the failure probability of the pipeline according to the frequency approximate probability theory.

[0012] According to a multi-defect pipeline evaluation method based on multi-dimensional joint distribution provided by the present invention, a multi-dimensional joint probability distribution model between defects is constructed according to the probability distribution of each defect and the relationship between the Vine Copula function and the reliability of the series pipeline system, which specifically includes: selecting a Vine Copula function, and selecting an optimal binary Copula function form for each variable pair according to the selected Vine Copula function; the variable pair is a combination of different defects that affect the failure state of the pipeline; after determining the binary Copula function form, estimating relevant parameters of the Copula function; based on the selected VineCopula structure, the binary Copula function form and its parameter estimation results, taking each defect probability distribution as an input marginal distribution, and constructing a multi-dimensional joint probability distribution model between defects.

[0013] According to a multi-defect pipeline evaluation method based on multi-dimensional joint distribution provided by the present invention, the reliability of the pipeline is determined according to the failure probability of the pipeline under the action of the single defect based on the multi-dimensional joint probability distribution model between defects, specifically comprising: based on the constructed multi-dimensional joint probability distribution model between defects, the joint failure probability of the pipeline under the action of multiple defects is determined according to the failure probability of the pipeline under the action of a single defect; using a survival Copula function, the joint failure probability is converted into the reliability of the pipeline; the reliability of the pipeline reflects the ability of the pipeline to maintain a safe state under the joint action of multiple defects.

[0014] According to a multi-defect pipeline evaluation method based on multi-dimensional joint distribution provided by the present invention, the estimation of the relevant parameters of the Copula function specifically includes: using the Spearman rank correlation coefficient as an evaluation index of the correlation measure to examine the nonlinear correlation and tail correlation between defects; deriving the relevant parameters of the Copula function based on the relationship between the Spearman rank correlation coefficient and the Copula function; the relevant parameters are used to describe the degree of correlation between different defects and are the key input for constructing a multi-dimensional joint probability distribution model between defects.

[0015] The present invention also provides a multi-defect pipeline evaluation device based on multi-dimensional joint distribution, comprising the following modules:

[0016] The probability distribution model of the pipeline under the action of a single defect is used to obtain input data, input the input data into the constructed probability distribution model of the pipeline under the action of a single defect, and output the failure probability of each defect of the pipeline and the probability distribution of each defect; the constructed probability distribution model of the pipeline under the action of a single defect is constructed according to actual pipeline detection data, historical failure data and pipeline defect-related evaluation standards;

[0017] A defect multi-dimensional joint probability distribution model construction module is used to construct a defect multi-dimensional joint probability distribution model according to the defect probability distribution and based on the relationship between the Vine Copula function and the reliability of the series pipeline system;

[0018] The reliability determination module is used to determine the reliability of the pipeline according to the failure probability of the pipeline under the action of the single defect based on the multi-dimensional joint probability distribution model between the defects.

[0019] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for evaluating a pipeline containing multiple defects based on multi-dimensional joint distribution as described above is implemented.

[0020] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for evaluating a pipeline containing multiple defects based on multi-dimensional joint distribution as described in any one of the above is implemented.

[0021] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements any of the above-mentioned methods for evaluating a pipeline containing multiple defects based on multi-dimensional joint distribution.

[0022] The present invention provides a method, device and equipment for evaluating a pipeline with multiple defects based on multi-dimensional joint distribution, including the following beneficial effects: by constructing a probability distribution model of the pipeline under the action of a single defect and considering the correlation between each defect, the reliability of the pipeline with multiple defects can be more accurately evaluated. Compared with the traditional method, the present invention avoids the problem of overestimation or underestimation of reliability due to ignoring the correlation between defects. By introducing the Vine Copula theory and constructing a multi-dimensional joint probability distribution model between defects that characterizes the reliability of the pipeline, the evaluation process can more comprehensively reflect the actual failure state of the pipeline system, providing a more refined reliability evaluation method, which helps to more accurately identify the potential risks of the pipeline. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0024] Figure 1 It is one of the flow charts of the pipeline evaluation method containing multiple defects based on multi-dimensional joint distribution provided by the present invention.

[0025] Figure 2 It is a schematic diagram of the decomposition of the four-dimensional Vine Copula structure provided by the present invention.

[0026] Figure 3 It is a structural schematic diagram of a pipeline evaluation device containing multiple defects based on multi-dimensional joint distribution provided by the present invention.

[0027] Figure 4 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are 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.

[0029] In recent years, in the field of energy storage and transportation, pipeline transportation is the most widely used and safe mode of transportation. Relevant data show that by the end of 2020, the total mileage of my country's oil and gas pipelines has reached 165,000 kilometers, including 31,000 kilometers of crude oil pipelines, 32,000 kilometers of refined oil pipelines, and 102,000 kilometers of natural gas pipelines. The national oil and gas pipelines have basically formed a network. However, the safety issues of oil and gas pipeline operation are increasing. During the operation of the pipelines, they are subjected to the combined effects of a variety of different failure mechanisms, such as corrosion, wear, impact and fatigue, resulting in frequent leakage accidents. The main cause of pipeline failure is the combined effect of various defects in the pipeline. Therefore, according to the basic data of the pipeline and each defect, it is particularly important to study the multidimensional joint probability distribution of pipeline system failure based on the correlation between defects and evaluate the safety of in-service pipelines.

[0030] The failure process of a pipeline system with multiple defects is often not caused by a single failure process. On the contrary, a pipeline system with multiple defects is caused by the combined action of multiple different failure modes, which in turn leads to failure. There is a certain correlation between different failure modes. Ignoring the correlation between them will reduce the accuracy of the reliability assessment results. The measurement of the correlation between failure modes is the focus and difficulty in reliability research. Before studying the correlation between multiple failure modes, researchers usually evaluate the reliability of systems with multiple failure modes according to the "weakest link principle" or "independence assumption principle". The weakest link principle assumes that each failure mode is completely correlated, the reliability of the system is determined by the weakest part in the system, and the system reliability is equal to the smallest of the reliabilities of each subsystem. The independence assumption principle assumes that each failure mode is independent of each other, and the reliability of the system is equal to the product of the reliabilities of each subsystem. This method is also a traditional serial reliability method. The weakest link principle and the independence assumption principle are both extreme assumption methods. The former often overestimates the reliability and the latter underestimates the reliability. The reliability assessment results obtained by considering the failure correlation are most in line with actual needs. Therefore, domestic and foreign scholars have done a lot of work on the reliability research related to multiple failure modes.

[0031] Xie mainly introduced the multi-component dependent failure model to study the impact of pipe segment failure on the failure probability of the pipeline system. The multi-component dependent failure model was used to estimate the probability of leakage failure or rupture of the pipeline system caused by the simultaneous leakage of multiple pipe segments, and an improved pipeline system failure probability model was proposed. Qijan et al. proposed a probabilistic method for evaluating the probability of failure of corroded pipelines based on the corrosion module of the FITNET FFS program, which takes into account the correlation between input variables. Zhou et al. proposed a method for evaluating the reliability of a time-varying system of a corroded pipeline segment containing multiple spatially associated active corrosion defects. Considering three different failure modes at a given corrosion defect, the pipeline segment was modeled as a series system. Stewart et al. developed an effective implicit limit state function based on the reliability method of probabilistic finite elements, which solved the problem of a high degree of conservatism when considering the interaction of defects in pipelines containing multiple defects and using a closed form explicit limit state function for probability evaluation. The above-mentioned partial probability methods have limitations in the reliability evaluation of pipelines containing multiple types of defects. Some methods consider the correlation between each defect by describing the correlation between input variables, but none of them considers the multi-dimensional joint probability distribution of the failure of the pipeline system containing multiple defects. The present invention proposes a method that takes the probability distribution of the pipeline limit state equation corresponding to each defect as a random variable, and further considers the correlation between defects from the defect effect result, that is, the pipeline structure state level compared to the input variable. The Vine Copula theory is introduced, the pipeline section is modeled as a series system to analyze its reliability, and a multi-dimensional joint probability distribution model between defects that characterizes the pipeline reliability is constructed to complete the reliability evaluation of the pipeline containing multiple defects.

[0032] The purpose of the present invention is to solve the problem that it is difficult to construct a multi-dimensional joint probability distribution of pipeline system failure based on the nonlinear correlation between defects of an oil and gas pipeline containing multiple defects, and to provide a reference for the daily maintenance of pipeline sections by introducing the Vine Copula theory to construct a multi-dimensional joint probability distribution model between defects that characterizes pipeline reliability.

[0033] Combine the following Figure 1-Figure 4 Embodiments of the present invention are described in detail.

[0034] The implementation of the present invention is mainly divided into three parts:

[0035] (1) The input of the scheme of the present invention is clarified: the probability distribution characteristics of the random variables of the pipeline body and defect size, and the pipeline limit state function (Z1, Z2, ..., Z2) corresponding to each defect is determined according to the relevant evaluation standards of pipeline defects. n );

[0036] (2) Based on the input data, the Monte Carlo method is used to construct a probability distribution (CDF) model of the pipeline under the action of a single defect, and the CDF (Cumulative Distribution Function) of the pipeline is output, that is, F1(z1),…,F n (z n ) and the failure probability P of the pipeline under the action of a single defect i =P(Z i <0)(i=1,2,…,n);

[0037] (3) According to the probability distribution of pipeline F1(z1),…,F n (z n ), based on the relationship between the Vine Copula function and the reliability of the series pipeline system, a multi-dimensional joint probability distribution model between defects that characterizes the pipeline reliability is constructed, and finally the reliability of the pipeline with multiple defects is output.

[0038] Figure 1 This is one of the flow charts of the multi-defect pipeline evaluation method based on multi-dimensional joint distribution provided by the present invention, such as Figure 1 As shown, the method comprises the following steps:

[0039] S110, obtaining input data, inputting the input data into the constructed probability distribution model of the pipeline under the action of a single defect, and outputting the failure probability of each defect of the pipeline and the probability distribution of each defect.

[0040] The constructed probability distribution model of the pipeline under the action of a single defect is constructed based on actual pipeline inspection data, historical failure data and pipeline defect-related evaluation standards.

[0041] According to a multi-defect pipeline evaluation method based on multi-dimensional joint distribution provided by the present invention, the input data is obtained, the input data is input into a constructed probability distribution model of a pipeline under a single defect, and the failure probability of each defect of the pipeline and the probability distribution of each defect are output, which specifically includes: obtaining basic random variables that affect the result state of the pipeline; obtaining the probability distribution characteristics of the basic random variables as input data; inputting the input data into the pipeline limit state equation to obtain an output result; the pipeline limit state equation is determined based on the evaluation standard related to pipeline defects; judging whether the execution condition of the failure probability model of the pipeline under a single defect is met according to the output result of the pipeline limit state equation; if the execution condition is met, then based on the failure probability model of the pipeline under a single defect, the failure probability of the pipeline under a single defect is determined according to the input data; the failure probability of the pipeline under the effect of each defect is obtained one by one, and the probability distribution of each defect with the pipeline limit state equation as a random variable is further fitted according to the failure probability of each defect.

[0042] According to a multi-dimensional joint distribution-based pipeline evaluation method provided by the present invention, the failure probability model of the pipeline under the action of a single defect is based on the input data to determine the failure probability of the pipeline under the action of a single defect, which specifically includes: using the Monte Carlo method to generate corresponding random variables according to the probability distribution characteristics of multiple basic random variables as sample data of the solution; substituting the sample data into the limit state equation of the pipeline, based on the law of large numbers, counting the number of failures, and obtaining the failure probability of the pipeline according to the frequency approximate probability theory.

[0043] Specifically, the limit state equation is introduced below.

[0044] According to the definition of limit state in reliability theory, when the entire structure or a part of the structure exceeds a certain state, that is, it cannot meet a certain functional requirement specified in the design, the specific state is the limit state. According to this definition, the functional function representing the state of the pipeline structure, also known as the limit state equation of the pipeline, is expressed as:

[0045] Z = g(X) = g(X1,X2,…,X n ) (1)

[0046] Where: X=(X1,X2,…,X n ) Tare n basic random variables that affect the pipeline result state, such as the basic parameters of the pipeline body (wall thickness, outer diameter, internal pressure, etc.) and defect size parameters (defect depth, defect length, etc.); Z is a random variable that characterizes the structural safety margin. When Z>0, the pipeline is in a safe state, when Z=0, the pipeline is in a limit state, and when Z<0, the pipeline is in a failure state. The pipeline limit state equations corresponding to different defects can be determined according to the relevant evaluation standards for pipeline defects.

[0047] The Monte Carlo method is introduced below.

[0048] From formula (1), we can see that the pipeline fails when Z<0. Therefore, the failure probability of the pipeline is P f The following formula can be used for calculation:

[0049]

[0050] Where: f X (x) = f X (x1,x2,…,x n ) is the joint probability density function of the pipeline basic random variables X; F X (x) = F X (x1,x2,…,x n ) is the joint cumulative distribution function of the pipeline basic random variables X.

[0051] Since the joint probability density function of the random variables of the pipeline in this engineering problem is difficult to obtain and the implementation of multiple integrals is difficult, it is not practical to directly calculate the failure probability using formula (2). The Monte Carlo method is one of the most commonly used methods to replace multiple integrals to calculate the failure probability. This method is a repeated simulation process. Each simulation uses the corresponding probability distribution of n basic random variables to generate the corresponding random variables and obtain the sample X of the solution. j =(X 1j ,X 2j ,…,X nj ) T (j=1,2,…,m), substitute the sample data into the limit state equation of the pipeline, and based on the law of large numbers, statistically analyze the number of failures and obtain the failure probability of the pipeline according to the frequency approximate probability theory. Through fitting, the probability distribution F1(z1),…,F with the pipeline limit state equation as the random variable is further obtained. n (z n ), most of the distribution forms are normal distribution. Among them, the probability distribution forms of pipeline random variables mainly include normal distribution, lognormal distribution, extreme value distribution and exponential distribution, which can be fitted and optimized according to the internal detection data of different pipelines.

[0052] S120. According to the probability distribution of each defect, based on the relationship between the Vine Copula function and the reliability of the series pipeline system, a multi-dimensional joint probability distribution model between defects is constructed.

[0053] According to a multi-defect pipeline evaluation method based on multi-dimensional joint distribution provided by the present invention, a multi-dimensional joint probability distribution model between defects is constructed according to the probability distribution of each defect and the relationship between the Vine Copula function and the reliability of the series pipeline system, which specifically includes: selecting a Vine Copula function, and selecting an optimal binary Copula function form for each variable pair according to the selected Vine Copula function; the variable pair is a combination of different defects that affect the failure state of the pipeline; after determining the binary Copula function form, estimating relevant parameters of the Copula function; based on the selected VineCopula structure, the binary Copula function form and its parameter estimation results, taking each defect probability distribution as an input marginal distribution, and constructing a multi-dimensional joint probability distribution model between defects.

[0054] According to a multi-defect pipeline evaluation method based on multi-dimensional joint distribution provided by the present invention, the estimation of the relevant parameters of the Copula function specifically includes: using the Spearman rank correlation coefficient as an evaluation index of the correlation measure to examine the nonlinear correlation and tail correlation between defects; deriving the relevant parameters of the Copula function based on the relationship between the Spearman rank correlation coefficient and the Copula function; the relevant parameters are used to describe the degree of correlation between different defects and are the key input for constructing a multi-dimensional joint probability distribution model between defects.

[0055] Specifically, the Vine Copula theory and its application in the present invention are introduced below.

[0056] Regarding the reliability evaluation of pipelines with multiple defects, the key point is that pipeline failure is the result of the combined effect of multiple defects. Constructing a multi-dimensional joint probability distribution of the pipeline system with multiple defects can more clearly reflect the failure state of the pipeline. Therefore, based on the present invention, the probability distribution of the pipeline limit state equation corresponding to each defect is considered as a random variable. The problem can be transformed into how to obtain the multi-dimensional joint probability distribution of the entire pipeline based on the failure probability distribution of each defect. In recent years, the Copula function has become an effective tool for modeling the correlation between random variables and constructing multi-dimensional joint probability distributions. It can be regarded as a connection function between marginal distributions and joint distributions. This technology can solve the difficulty of the reliability evaluation problem of pipelines with multiple defects. Therefore, the present invention introduces the Copula theory, takes the pipeline limit state equation corresponding to each defect as a random variable, and uses the Copula function to describe the correlation between the defects. At the same time, the probability distribution of each defect F1(z1),…,F obtained based on the above-mentioned Monte Carlo method is n (z n ), and construct a high-dimensional joint probability distribution model to characterize the reliability and failure probability of the pipeline. The following will describe the implementation process in detail from several aspects.

[0057] (1) Vine Copula Theory

[0058] Given the marginal distribution F1(x1),…,F n (x n ), let u i =F i (x i ), solve its multidimensional joint distribution F(x1,…,x n ), according to Sklar's theorem, there exists a Copula function that satisfies:

[0059] F(x1,…,x n )=C{F1(x1),…,F n (x n )}=C(u1,…,u n ) (3)

[0060] The multivariate distribution is decomposed into its marginal distribution and a Copula function, which fully describes the interrelated structure of the marginal distribution. Table 1 shows the basic properties of the Copula theory. Introducing this theory, the present invention converts each defect probability distribution F1(z1),…,F n (z n ) as the input marginal distribution of the Copula function to construct the multidimensional joint probability distribution F(z1,…,z n ).

[0061] Table 1 Basic properties of Copula theory

[0062]

[0063] Since it is difficult to obtain relevant parameters in the process of constructing high-dimensional joint probability distribution using Copula function, and the binary Copula function is relatively simple, and when constructing the joint distribution function of multi-dimensional related variables, it is also necessary to consider the correlation problem of multi-dimensional variables. Therefore, in order to solve the above problems, researchers proposed the idea of ​​constructing paired Copula modeling (Pair-Copula Construction, PCC), and continuously developed the Regular Vine Copula model theory. Based on the Regular Vine Copula model theory, the joint distribution function of a complex pipeline system can be decomposed into the product of multiple paired Copulas and the CDF of the corresponding defect limit state function. The probability density function PDF of n-dimensional PCC is:

[0064]

[0065] Where: And i<j, i1<i2<…<i k . Obviously, there is not only one way to decompose.

[0066] Bedford and Cooke proposed a graph structure called Regular Vine to visualize the process of constructing multidimensional distributions. The definition of Regular Vine: An n-dimensional vine is a sequence of n-1 trees with the following properties: (1) Tree j has n+1-j nodes and nj edges; (2) The edges in tree j are nodes in tree j+1; (3) The proximity condition: if the corresponding edges in tree j share a node, then two nodes in tree j+1 are connected by an edge. According to this definition, the joint density represented by the vine is given by the density of the entire graph. The density of the Pair-Copula (binary copula) on the edges and the edge density of n nodes in the first-level tree. Among the many different PCC methods, there are two special types of Regular Vine: Canonical Vine (C-vine) and Stretchable Vine (D-vine). For C-vine, each tree has a unique node connected to all other nodes. For D-vine, each tree is a path. Figure 1 The schematic diagrams of the C-vine and D-vine structural decompositions of the four variables are shown respectively. The joint PDFf(x1,x2,x3,x4) and CDFF(x1,x2,x3,x4) of C-vine and D-vine are shown below.

[0067]

[0068]

[0069] like Figure 2 The figure shows the schematic diagram of the decomposition of the four-dimensional Vine Copula structure. In the figure, T1, T2, and T3 are the three trees of the four-dimensional C-vine Copula and D-vine Copula respectively. Figure 2 It can be seen that each tree in the C-vine structure has a main variable connected to the remaining variables, and the two types of combinations are merged layer by layer to finally obtain the joint probability function of the multidimensional random variables. Different selections of main variables in the C-vine structure will lead to model diversification. In practical applications, the correlation between each random variable is often calculated, and the variable with the largest sum of correlation coefficients with other variables is selected as the main variable. The Person correlation coefficient, Kendall Rank correlation coefficient or Spearman Rank correlation coefficient are often used to describe the correlation between variables. There is no primary and secondary relationship between the variables in the D-vine structure, and the nodes of each tree are connected in sequence. Generally speaking, when there is a main variable that guides other variables in the random variable, the C-vine structure is selected to describe the variable, and when the random variables are independent of each other, the D-vine structure is selected to describe the variable.

[0070] There are many types of binary copula functions in Vine Copula. Currently, the two most commonly used types of Copula functions are elliptic Copula and Archimedean Copula. Among them, Gaussian Copula and binary Student t Copula are elliptic Copula functions. The two marginal distributions of these two functions satisfy the characteristics of origin symmetry, and Student t Copula often introduces the degree of freedom parameter α2. When α2→∞, Student t Copula will be close to Gaussian Copula. Archimedean Copula is a function cluster constructed by a "generating function", and the commonly used ones are Frank Copula, Clayton Copula and Gumbel Copula. The present invention intends to use these 5 types of Copula functions to establish the joint probability distribution between defects, and the function form is shown in Table 2.

[0071] Table 2 Proposed Copula function form

[0072]

[0073] (2) Vine Copula function form selection and parameter estimation

[0074] After determining the structure of the Vine Copula function, it is necessary to select the Copula function family of each binary Copula function, and use the Spearman Rank correlation coefficient to estimate the parameters of the Vine Copula function to determine the multidimensional joint distribution function that represents the correlation between multiple defects in the pipeline.

[0075] □ Select the binary conditional Copula function forms that make up the Vine Copula. Use the information criterion quantitative calculation method to evaluate the Copula function form selected from the two perspectives of goodness of fit and complexity. Commonly used information criterion quantitative calculation methods include Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC). The AIC calculation formula is formula (5), and the BIC calculation formula is formula (6).

[0076] AIC=2k-2ln(L) (5)

[0077] BIC=-2ln(L)+kln(N) (6)

[0078] Where: k is the number of parameters in the Copula function; L is the value of the maximum likelihood function of the Copula function; N is the sample size. The AIC information criterion focuses more on selecting a probability model that fits the existing samples better, while the BIC criterion takes into account the number of samples. When the number of samples is too large, it can effectively prevent the model complexity from being too high due to excessive model accuracy. In the implementation scheme of the present invention, more emphasis is placed on selecting a suitable model to describe the sample distribution, so the binary conditional Copula type with the smallest AIC is selected for parameter estimation.

[0079] □For the Copula function, the maximum likelihood method can generally be used to estimate the relevant parameters α of the Copula function. However, since the probability distribution of the input variable, i.e., the defect, and the corresponding parameters of the link function Copula function must be estimated jointly, this method will bring computational burden. Compared with the maximum likelihood method, when the Spearman Rank correlation coefficient ρ is known, s Under the premise of s The correlation parameter α is derived from the relationship between the copula and the Spearman Rank correlation coefficient.

[0080] The Spearman Rank correlation coefficient replaces the original variable value with the corresponding order of the values ​​in the variable. In fact, it replaces X and Y with the Pearson correlation coefficient of their corresponding levels x and y. The specific formula is as follows:

[0081]

[0082] Compared with the Pearson correlation coefficient, which is only applicable to describing the degree of correlation between random variables with linear relationship, continuity, and normal distribution, the Spearman Rank correlation coefficient can better describe the monotonic correlation between two variables and is less sensitive to significant outliers in the sample.

[0083] In order to examine the nonlinear correlation measure and tail correlation measure between defects, the Spearman Rank correlation coefficient is used as the evaluation index of the correlation measure. The Spearman Rank correlation coefficient has the following relationship with the Copula function:

[0084]

[0085] Where: s represents the Spearman Rank correlation coefficient between variables; α represents the correlation parameter of the corresponding Copula function; u and v represent two variables; C(·) represents the Copula function.

[0086] S130. Based on the multi-dimensional joint probability distribution model between defects, determine the reliability of the pipeline according to the failure probability of the pipeline under the action of the single defect.

[0087] According to a multi-defect pipeline evaluation method based on multi-dimensional joint distribution provided by the present invention, the reliability of the pipeline is determined according to the failure probability of the pipeline under the action of the single defect based on the multi-dimensional joint probability distribution model between defects, specifically comprising: based on the constructed multi-dimensional joint probability distribution model between defects, the joint failure probability of the pipeline under the action of multiple defects is determined according to the failure probability of the pipeline under the action of a single defect; using a survival Copula function, the joint failure probability is converted into the reliability of the pipeline; the reliability of the pipeline reflects the ability of the pipeline to maintain a safe state under the joint action of multiple defects.

[0088] Specifically, (3) pipeline system reliability.

[0089] For the action process of a single defect i, the corresponding limit state equation Z i Satisfy {Z i <0}, it is judged as a pipeline failure event, P i =P(Z i <0) is the failure probability of the pipeline under the effect of this defect, corresponding to Z i The cumulative failure probability (CDF) of the pipeline is i =P(Z i ≥0)=1-P i Therefore, the reliability of the series system with multiple defective pipelines is:

[0090]

[0091] Where: represents the survival Copula function, as shown in Table 1.

[0092] The present invention proposes a method of using the probability distribution of the pipeline limit state equation corresponding to each defect as a random variable. Compared with considering the correlation of input variables, the correlation between defects is further considered from the defect effect result, that is, the pipeline structure state level.

[0093] The Vine Copula theory is introduced to model the pipeline section as a series system to analyze its reliability, and a multi-dimensional joint probability distribution model between defects that characterizes the pipeline reliability is constructed to complete the reliability evaluation of pipelines with multiple defects. Compared with the traditional series reliability independence principle, which is more in line with reality, the multi-dimensional joint probability distribution of the pipeline system more clearly reflects the failure state of the pipeline.

[0094] The following is a description of the pipeline evaluation device containing multiple defects based on multidimensional joint distribution provided by the present invention. The pipeline evaluation device containing multiple defects based on multidimensional joint distribution described below and the pipeline evaluation method containing multiple defects based on multidimensional joint distribution described above can be referenced to each other.

[0095] like Figure 3 The present invention provides a multi-defect pipeline evaluation device based on multi-dimensional joint distribution, comprising:

[0096] The probability distribution model 310 of the pipeline under the effect of a single defect is used to obtain input data, input the input data into the constructed probability distribution model of the pipeline under the effect of a single defect, and output the failure probability of each defect of the pipeline and the probability distribution of each defect; the constructed probability distribution model of the pipeline under the effect of a single defect is constructed according to actual pipeline inspection data, historical failure data and pipeline defect-related evaluation standards;

[0097] A defect multi-dimensional joint probability distribution model construction module 320 is used to construct a defect multi-dimensional joint probability distribution model according to the defect probability distribution and based on the relationship between the Vine Copula function and the reliability of the series pipeline system;

[0098] The reliability determination module 330 is used to determine the reliability of the pipeline according to the failure probability of the pipeline under the single defect based on the multi-dimensional joint probability distribution model between defects.

[0099] Figure 4 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 4As shown, the electronic device may include: a processor 410, a communication interface 420, a memory 430 and a communication bus 440, wherein the processor 410, the communication interface 420 and the memory 430 communicate with each other through the communication bus 440. The processor 410 may call the logic instructions in the memory 430 to execute the evaluation method of a pipeline containing multiple defects based on a multi-dimensional joint distribution, the method comprising: obtaining input data, inputting the input data into a constructed probability distribution model of a pipeline under a single defect, and outputting the failure probability of each defect of the pipeline and the probability distribution of each defect; the constructed probability distribution model of the pipeline under a single defect is constructed according to actual pipeline detection data, historical failure data and pipeline defect-related evaluation standards; according to the probability distribution of each defect, based on the relationship between the Vine Copula function and the reliability of the series pipeline system, a multi-dimensional joint probability distribution model between defects is constructed; based on the multi-dimensional joint probability distribution model between defects, the reliability of the pipeline is determined according to the failure probability of the pipeline under a single defect.

[0100] In addition, the logic instructions in the above-mentioned memory 430 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0101] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the multi-defect pipeline evaluation method based on multi-dimensional joint distribution provided by the above methods, the method including: obtaining input data, inputting the input data into a constructed probability distribution model of the pipeline under the action of a single defect, and outputting the failure probability of each defect of the pipeline and the probability distribution of each defect; the constructed probability distribution model of the pipeline under the action of a single defect is constructed according to actual pipeline detection data, historical failure data and pipeline defect-related evaluation standards; according to the probability distribution of each defect, based on the relationship between the Vine Copula function and the reliability of the series pipeline system, a multi-dimensional joint probability distribution model between defects is constructed; based on the multi-dimensional joint probability distribution model between defects, the reliability of the pipeline is determined according to the failure probability of the pipeline under the action of a single defect.

[0102] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented when the computer program is executed by a processor to execute the pipeline evaluation method containing multiple defects based on multi-dimensional joint distribution provided by the above-mentioned methods, the method comprising: obtaining input data, inputting the input data into a constructed probability distribution model of the pipeline under the action of a single defect, and outputting the failure probability of each defect of the pipeline and the probability distribution of each defect; the constructed probability distribution model of the pipeline under the action of a single defect is constructed according to actual pipeline detection data, historical failure data and pipeline defect-related evaluation standards; according to the probability distribution of each defect, based on the relationship between the Vine Copula function and the reliability of the series pipeline system, a multi-dimensional joint probability distribution model between defects is constructed; based on the multi-dimensional joint probability distribution model between defects, the reliability of the pipeline is determined according to the failure probability of the pipeline under the action of a single defect.

[0103] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0104] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0105] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A pipeline evaluation method with multiple defects based on multi-dimensional joint distribution, characterized in that: include: Acquire input data, input the input data into a constructed probability distribution model of a pipeline under a single defect, and output the failure probability of each defect of the pipeline and the probability distribution of each defect; the constructed probability distribution model of the pipeline under a single defect is constructed according to actual pipeline inspection data, historical failure data and pipeline defect-related evaluation standards; According to the probability distribution of each defect, based on the relationship between the Vine Copula function and the reliability of the series pipeline system, a multi-dimensional joint probability distribution model between defects is constructed; Based on the multi-dimensional joint probability distribution model between defects, the reliability of the pipeline is determined according to the failure probability of the pipeline under the action of the single defect.

2. The method for evaluating a pipeline with multiple defects based on multi-dimensional joint distribution according to claim 1 is characterized in that: The obtaining of input data, inputting the input data into the constructed probability distribution model of the pipeline under the action of a single defect, and outputting the failure probability of each defect of the pipeline and the probability distribution of each defect specifically include: Get the basic random variables that affect the status of the pipeline results; Obtaining probability distribution characteristics of the basic random variable as input data; Inputting the input data into a pipeline limit state equation to obtain an output result; the pipeline limit state equation is determined based on a pipeline defect-related evaluation standard; Judging whether the execution condition of the failure probability model of the pipeline under the action of a single defect is satisfied according to the output result of the pipeline limit state equation; If the execution condition is met, then based on the failure probability model of the pipeline under the single defect, the failure probability of the pipeline under the single defect is determined according to the input data; The failure probability of the pipeline under the action of each defect is obtained one by one, and the probability distribution of each defect with the pipeline limit state equation as the random variable is further fitted according to the failure probability of each defect.

3. The method for evaluating a pipeline with multiple defects based on multi-dimensional joint distribution according to claim 2 is characterized in that: The failure probability model of the pipeline under the effect of a single defect is based on the input data to determine the failure probability of the pipeline under the effect of a single defect, specifically including: Through the Monte Carlo method, the corresponding random variables are generated according to the probability distribution characteristics of multiple basic random variables as sample data of the solution; Substitute the sample data into the limit state equation of the pipeline, count the number of failures based on the law of large numbers, and obtain the failure probability of the pipeline according to the frequency approximate probability theory.

4. The method for evaluating a pipeline with multiple defects based on multi-dimensional joint distribution according to claim 1, characterized in that: According to the probability distribution of each defect, based on the relationship between the Vine Copula function and the reliability of the series pipeline system, a multi-dimensional joint probability distribution model between defects is constructed, which specifically includes: A Vine Copula function is selected, and based on the selected Vine Copula function, an optimal binary Copula function form is selected for each variable pair; the variable pair is a combination of different defects that affect the failure state of the pipeline; After determining the form of the binary Copula function, estimate the relevant parameters of the Copula function; Based on the selected Vine Copula structure, binary Copula function form and parameter estimation results, the probability distribution of each defect is taken as the input marginal distribution to construct a multidimensional joint probability distribution model among defects.

5. The method for evaluating a pipeline with multiple defects based on multi-dimensional joint distribution according to claim 1, characterized in that: The method of determining the reliability of the pipeline based on the multi-dimensional joint probability distribution model between defects and the failure probability of the pipeline under the action of the single defect specifically includes: Based on the constructed multi-dimensional joint probability distribution model between defects, the joint failure probability of the pipeline under the action of multiple defects is determined according to the failure probability of the pipeline under the action of a single defect; Using the survival Copula function, the joint failure probability is converted into the reliability of the pipeline; The reliability of the pipeline reflects the ability of the pipeline to maintain a safe state under the combined effects of multiple defects.

6. The method for evaluating a pipeline with multiple defects based on multi-dimensional joint distribution according to claim 4 is characterized in that: The estimated Copula function parameters specifically include: The Spearman rank correlation coefficient is used as the evaluation index of correlation measure to examine the nonlinear correlation and tail correlation between defects; According to the relationship between the Spearman rank correlation coefficient and the Copula function, the relevant parameters of the Copula function are derived; The correlation parameters are used to describe the degree of correlation between different defects and are key inputs for constructing a multi-dimensional joint probability distribution model between defects.

7. A pipeline evaluation device containing multiple defects based on multi-dimensional joint distribution, characterized in that: include: The probability distribution model of the pipeline under the action of a single defect is used to obtain input data, input the input data into the constructed probability distribution model of the pipeline under the action of a single defect, and output the failure probability of each defect of the pipeline and the probability distribution of each defect; the constructed probability distribution model of the pipeline under the action of a single defect is constructed according to actual pipeline detection data, historical failure data and pipeline defect-related evaluation standards; A module for constructing a multi-dimensional joint probability distribution model between defects, for constructing a multi-dimensional joint probability distribution model between defects based on the probability distribution of each defect and the relationship between the VineCopula function and the reliability of the series pipeline system; The reliability determination module is used to determine the reliability of the pipeline according to the failure probability of the pipeline under the action of the single defect based on the multi-dimensional joint probability distribution model between the defects.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method for evaluating a pipeline containing multiple defects based on multi-dimensional joint distribution as described in any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for evaluating a pipeline containing multiple defects based on multi-dimensional joint distribution as described in any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for evaluating a pipeline containing multiple defects based on multi-dimensional joint distribution as described in any one of claims 1 to 6 is implemented.

Images