TrFISM-MICMAC method for severity analysis of underwater manifold leakage accident

Abstract

The invention belongs to the technical field of underwater manifold leakage accident prevention and control, and discloses a TrFISM-MICMAC method for severity analysis of an underwater manifold leakage accident. The method comprises the following steps: firstly, constructing a comprehensive influence matrix of each risk source of the underwater manifold leakage accident, obtaining a reachable matrix representing the mutual relation between the risk sources through iterative calculation, then based on the multiplication of a trapezoidal fuzzy interpretation structure model and a cross matrix, converting an analysis result into a visual chart, comparing the influence priorities of each risk source, and determining the risk source of the underwater manifold leakage accident. And finally, according to an investigation statistical result, based on a weighted membership degree method, calculating to obtain a severity index of each risk factor. The theoretical framework has flexibility and universality, and can be suitable for leakage risk analysis of different types of underwater manifolds and other ocean engineering equipment. Meanwhile, the method has low requirements on data input and is convenient to popularize and apply in engineering practice.

Description

Technical Field

[0001] The present invention belongs to the technical field of underwater pipeline network leakage accident prevention and control, and relates to a TrFISM-MICMAC method for analyzing the severity of underwater pipeline network leakage accidents. Background Technique

[0002] The underwater pipeline network is a core component in the underwater oil and gas production system. Its main function is to achieve the convergence and diversion of fluids from multiple wells, and control the flow direction and flow rate of production fluids through a series of valves and pipelines. However, due to the complex working environment and long-term exposure to high pressure, high temperature, and corrosive fluids, the risk of leakage accidents in the underwater pipeline network cannot be ignored. Pipeline network leakage accidents may not only lead to serious waste of oil and gas resources, but also cause irreparable damage to the marine ecological environment, and even lead to production shutdowns and economic losses. Therefore, how to effectively evaluate the severity of underwater pipeline network leakage accidents and accurately identify the main risk sources has become a key technical problem in this field.

[0003] Existing studies mainly adopt traditional fuzzy comprehensive evaluation methods and accident analysis techniques based on statistics. Although these methods can qualitatively analyze risk sources, they have the following deficiencies when dealing with the complex interrelationships and influences among multiple risk sources:

[0004] (1) Insufficient analysis of the interaction relationships among risk sources. Traditional methods usually assume that each risk source exists independently, without fully considering the complex interaction relationships among risk sources. This simplified assumption may lead to an underestimation or overestimation of the influence of certain key risk sources, thereby affecting the reliability of the evaluation results.

[0005] (2) Insufficient visualization and intuitiveness of the analysis results. Existing technologies pay less attention to the visual expression of risk assessment results, resulting in the difficulty of intuitively presenting the influence priority and relative severity of each risk source in the analysis results, which limits the application of the evaluation results in actual decision-making.

[0006] The main reasons for the above deficiencies are that existing technologies have insufficient ability to handle the interrelationships of multiple factors in the theoretical framework, and lack technical means to integrate fuzzy theory and dynamic system analysis.

[0007] To address the above problems, the present invention proposes a method for analyzing the severity of underwater pipeline network leakage accidents based on the trapezoidal fuzzy interpretive structural model (TrFISM) and the cross matrix multiplication method (MICMAC). By constructing a comprehensive influence matrix of risk sources and iteratively calculating the reachability matrix, the dynamic interaction relationships among various risk sources are quantitatively analyzed; and based on the MICMAC method, the influence priority of each risk source is visualized, and its normalized weight is calculated. Combining the weighted membership degree method, the present invention can accurately evaluate the severity index of risk factors, providing a scientific and efficient technical solution for the leakage risk management of underwater pipeline networks. Summary of the Invention

[0008] The object of the present invention is to provide a TrFISM-MICMAC method for analyzing the severity of underwater pipeline leakage accidents, so as to solve the problems existing in the prior art.

[0009] The technical solution of the present invention is as follows:

[0010] A TrFISM-MICMAC method for analyzing the severity of underwater pipeline leakage accidents comprises the following steps:

[0011] Step 1: Based on the analytic hierarchy process and accident causation theory, adopt a three-level progressive model of "target layer - primary index - risk source"; starting from the target layer, disassemble the risk factors of underwater pipeline leakage accidents layer by layer from top to bottom through a vertical causal chain to establish a preliminary hierarchical relationship; among them, the target layer at the highest level is the risk of underwater pipeline leakage accidents, the primary indexes at the middle layer are the macro risk categories directly causing accidents (such as risks of equipment and materials, risks of operation and maintenance, etc.), and the risk sources at the bottom layer are the minimum risk units further disassembled from the primary indexes, that is, specific risk factors. The risk sources are mapped to the primary indexes through logical merging and finally contribute to the target layer risk;

[0012] Step 2: Invite domain experts to evaluate the influence degree between various risk factors and convert it into corresponding trapezoidal fuzzy numbers, and integrate and construct a comprehensive relationship matrix;

[0013] Step 3: Construct an adjacency matrix and a reachability matrix;

[0014] Step 4: Calculate the influence degree and dependence degree of each risk factor on the underwater pipeline system, and draw an influence-dependence relationship diagram of each risk factor;

[0015] Step 5: Calculate the reachable set and antecedent set of each risk factor;

[0016] Step 6: Calculate the priority of each risk factor for the underwater pipeline system, and obtain the severity weight vector of each risk factor in the underwater pipeline leakage accident after normalization;

[0017] Step 7: Through the fuzzy comprehensive evaluation method, quantitatively calculate the severity of the risk sources in the underwater pipeline leakage accident.

[0018] The present invention first constructs a comprehensive influence matrix of various risk sources for underwater pipeline network leakage accidents, and obtains an accessible matrix representing the mutual relationship between risk sources through iterative calculation. Then, based on the trapezoidal fuzzy interpretive structural model (TrFISM) and the multiplication method of the cross matrix (MICMAC), the analysis results are transformed into visual charts to compare the influence priorities of various risk sources and converted into normalized weights. Finally, according to the survey and statistical results, the severity index of each risk factor is calculated based on the weighted membership degree method.

[0019] The beneficial effects of the present invention are as follows:

[0020] (1) The present invention uses the TrFISM model to construct a comprehensive influence matrix of risk sources and generates an accessible matrix through iterative calculation, thereby systematically analyzing the multi-level interaction relationship between underwater pipeline network leakage risk sources, avoiding the limitations of the independence assumption of risk sources in traditional evaluation models, and being able to more comprehensively and truly reveal the dynamic association between various risk sources, providing a scientific basis for accurately identifying key risk factors;

[0021] (2) By combining the MICMAC method, the present invention transforms the interaction results between risk sources into quantified influence priorities and further normalizes them into dynamic weights, effectively making up for the deficiency of insufficient response to dynamic environmental changes in traditional risk weight allocation methods, making the risk assessment results more accurate and adaptable;

[0022] (3) The theoretical framework of the present invention is both flexible and general, and can be applied to the leakage risk analysis of different types of underwater pipeline networks and other marine engineering equipment. At the same time, the present invention has low requirements for data input and is convenient for popularization and application in engineering practice. Description of the Drawings

[0023] Figure 1 is the flow chart of the method for analyzing the severity of underwater pipeline network leakage accidents provided by the present invention;

[0024] Figure 2 is the influence-dependence diagram of various risk factors for underwater pipeline network leakage provided by the present invention;

[0025] Figure 3 is the diagram of the severity weights and evaluation results of underwater pipeline network leakage risk sources provided by the present invention. Detailed Embodiments

[0026] The specific structure and implementation process of the present solution will be described in detail below through specific embodiments and drawings.

[0027] As Figure 1 shown, in an embodiment of the present invention, a TrFISM-MICMAC method for analyzing the severity of underwater pipeline network leakage accidents is disclosed, including the following steps:

[0028] Step 1: Identify each risk factor of the underwater pipeline leakage accident and establish a preliminary hierarchical relationship, as shown in Table 1.

[0029] Table 1 Preliminary hierarchical structure model

[0030]

[0031] Step 2: Invite domain experts to conduct pairwise evaluations on the influence degrees between each risk factor. The conversion rules between the comment scales and the corresponding trapezoidal fuzzy numbers are shown in Table 2.

[0032] Table 2 Comment scales and conversion rules

[0033]

[0034] According to the expert evaluation results, use the mean operator to integrate the evaluation matrices of different experts into the initial relationship matrix M; according to the conversion rules in the above table, convert it into the corresponding trapezoidal fuzzy numbers to construct the fuzzy relationship matrix as follows:

[0035]

[0036] In the formula, is the trapezoidal fuzzy number representing the mutual influence degree between risk factor X i and X j ; in order to retain as much uncertain information in the trapezoidal fuzzy number as possible, introduce the α-cut set; the value of α ranges from 0 to 1 and is determined by the actual operator; when α = 1, it represents the maximum degree of uncertainty; when α = 0, it represents the minimum degree of uncertainty; when α = 0.5, it represents a neutral attitude; on this basis, establish the interval-valued relationship matrix Q as follows:

[0037]

[0038] In the formula, n represents the number of risk factors, represents the interval value of the element under the α value, and represent the minimum and maximum values of the α-cut set respectively, and the calculation methods are as follows:

[0039]

[0040] Perform defuzzification processing on the interval-valued relationship matrix Q through the following formula to obtain the comprehensive relationship matrix Z:

[0041]

[0042] Step 3, construct the adjacency matrix and reachability matrix.

[0043] Set a threshold θ, which is the mean value of the comprehensive relationship matrix Z; by comparing with the threshold θ, construct the adjacency matrix L = [l ij n×n , l ij represents the influence relationship between risk factors X i and X j . The value is determined by the following formula:

[0044]

[0045] Construct the reachability matrix K, as follows:

[0046]

[0047] In the formula, E represents the identity matrix equal to the size of the adjacency matrix, n is the number of risk factors, and at the same time represents the actual number of iterations, k ij is the element in the reachability matrix, representing the direct or indirect reachability relationship between risk factors X i and X j .

[0048] Step 4, calculate the influence degree and dependence degree of each factor on the whole system, and draw the influence-dependence relationship diagram of each factor.

[0049] First, calculate the influence degree and dependence degree of each risk factor X t on the underwater pipeline system. The calculation method is as follows:

[0050]

[0051] In the formula, DI t represents the influence degree of risk factor X t on the underwater pipeline system, and DD t represents the dependence degree of risk factor X t on the underwater pipeline system;

[0052] Then, draw the influence-dependence relationship diagram of each risk factor according to the influence degree DI and dependence degree DD of each risk factor; classify all risk factors according to the four-quadrant principle, into autonomous risks (A), dependent risks (B), chain risks (C) and independent risks (D), corresponding to the third quadrant, the fourth quadrant, the first quadrant and the second quadrant respectively.

[0053] Step 5, calculate the reachable set and antecedent set of each element.

[0054] First, based on the reachability matrix K, calculate the reachable set R of each risk factor through the following formula i ​With the antecedent set S i ; For risk factor X i , the reachable set R i contains all risk factors reachable from X i , and the antecedent set S i contains all risk factors that can reach X i ; Then, calculate the intersection T i between two risk elements to determine the top-level risk factor; Then, delete the rows and columns where the top-level risk element is located in the reachability matrix K, and calculate iteratively until all hierarchical structures and corresponding elements are determined;

[0055] R i = {X i | X i ∈ X, k ji ≠ 0}

[0056] S i = {X i | X i ∈ X, k ij ≠ 0}

[0057] T i = R i ∩ S i

[0058] In the formula, X i represents the i-th risk factor, and X represents the set of all risk factors of the underwater pipeline leak accident.

[0059] Step 6, calculate the priority of each factor for the entire system.

[0060] Calculate the priority D i of each risk factor through the following formula, and after normalization, obtain the severity weight vector w = (w1, w2,..., w n ) T ;

[0061] D i = DI i + DD i

[0062]

[0063] Step 7, through the fuzzy comprehensive evaluation method, quantitatively calculate the severity of the risk source in the underwater pipeline leak accident.

[0064] Establish a preliminary hierarchical structure for the analyzed risk factors of underwater pipeline leakage accidents. Take each risk factor as the factor set C, formulate severity evaluation indicators according to the 5-level scaling method, and obtain the alternative set V; use the weight vector w = (w1, w2, …, w n ) T obtained in step 6 as the weight set; distribute expert questionnaires and evaluate each risk factor according to the alternative set V; collect the questionnaires and establish a fuzzy matrix R:

[0065]

[0066] In the formula, r iq (i = 1, 2, …, n; q = 1, 2, 3, 4, 5) represents the membership degree value of the risk factor X t under the comment q;

[0067] Aggregate the weight set W and the fuzzy matrix R, and calculate the comprehensive index of the fuzzy matrix R under the influence of the weight through the following formula to obtain the fuzzy evaluation result vector P;

[0068]

[0069] Use the weighted average method to perform weighted calculations on each comment in the alternative set with the result vector P = (p1, p2, …, p5) as the weight, which is used as the severity evaluation result of the final underwater pipeline leakage accident.

[0070] In this embodiment, the trapezoidal fuzzy interpretive structural model (TrFISM) and the multiplication method of the cross matrix (MICMAC) are first used to analyze the degree of mutual influence between the risk sources of underwater pipeline leakage accidents, draw a hierarchical structure diagram and an influence-dependence relationship diagram, and classify all risk sources into four categories according to the degree of influence and dependence of each risk source relative to other risk sources to represent its influence in the overall system. Then, by comprehensively judging the influence and affected effects of the risk sources in the system, calculate their priorities, and normalize to obtain the weight values of each risk source. Finally, based on the expert investigation method and the fuzzy comprehensive evaluation method, determine the severity of the underwater pipeline leakage accident.

[0071] This embodiment has the following beneficial effects compared with the prior art:

[0072] (1) The present invention uses the TrFISM model to construct a comprehensive influence matrix of risk sources and generates a reachable matrix through iterative calculation, thereby systematically analyzing the multi-level interaction relationship between the risk sources of underwater pipeline leakage, avoiding the limitations of the independence assumption of risk sources in traditional evaluation models, and being able to more comprehensively and truly reveal the dynamic association between risk sources, providing a scientific basis for accurately identifying key risk factors;

[0073] (2) By combining the MICMAC method, the present invention converts the interaction results between risk sources into quantified influence priorities, and further normalizes them into dynamic weights, effectively making up for the defect of insufficient response to dynamic environmental changes in traditional risk weight allocation methods, making the risk assessment results more accurate and adaptable.

[0074] (3) The theoretical framework of the present invention is both flexible and general, and can be applied to the leakage risk analysis of different types of underwater manifolds and other offshore engineering equipment. At the same time, the present invention has low requirements for data input, which is convenient for popularization and application in engineering practice.

[0075] The following takes the underwater manifold arranged in a certain oil and gas field as an example, and combines the attached drawings to make a detailed description of the evaluation method of the present invention.

[0076] 1. Construct a preliminary risk hierarchy model for underwater manifold leakage accidents, as shown in Table 1. This model is divided into 3 layers and contains a total of 20 risk nodes, including 15 risk factors. By analyzing the mutual relationships of each risk factor, the severity of the overall system is calculated.

[0077] 2. Invite five experts in the field to evaluate the influence degree of the 15 risk nodes in the index layer in Table 1, and summarize the expert opinions to construct a fuzzy relation matrix. Then calculate the comprehensive relation matrix Z, and adopt a neutral attitude with α = 0.5, as shown in Table 3.

[0078] Table 3 Comprehensive relation matrix

[0079]

[0080] 3. In order to effectively distinguish the strength of the mutual influence between each node, take the threshold θ as the mean value of the comprehensive relation matrix Z, calculate the mutual relationships between each matrix, and summarize them into an adjacency matrix L, as shown in Table 4. Then calculate the reachability matrix K of all factors, and the results are shown in Table 5.

[0081] Table 4 Adjacency matrix

[0082]

[0083]

[0084] Table 5 Comprehensive reachability matrix

[0085]

[0086] 4. Based on the calculation results of the reachability matrix K, calculate the influence degree DI and dependence degree DD of each factor, and summarize the results in Table 6, and draw an influence-dependence relationship diagram according to the calculation results, as Figure 2 shown.

[0087] Summary Table of Influence Degree and Dependence Degree of Each Factor

[0088]

[0089]

[0090] All nodes are classified according to the four - quadrant principle into the following four risk factors:

[0091] (A) Autonomous risks. These risk factors are located in the third quadrant, with relatively low influence degree and dependence degree in the overall system, indicating that their interaction with other risk factors is weak, their own risk state is relatively stable, not easily disturbed by other risk factors, and not likely to affect other factors either. They include X7, X8, X9, X10.

[0092] (B) Dependent risks. These risk factors are located in the fourth quadrant, characterized by low influence degree but high dependence degree. Such risk factors are easily affected by other factors and change, and have a strong dependence on other factors. They include X1, X2, X3, X4, X6.

[0093] (C) Chain risks. These risk factors are located in the first quadrant, with relatively high influence degree and dependence degree, and poor stability in the overall system. They are easily affected by changes in other risk factors and are also likely to feedback their own changes to other factors, showing strong correlation, and need to be particularly emphasized in the risk management of FPSO fire and explosion. They include X13, X14.

[0094] (D) Independent risks. These risk factors are located in the second quadrant, with relatively high influence degree and low dependence degree. Therefore, when the state of such factors changes, they have a greater interference on other factors and are the key factors and main driving forces leading to system changes. They include X5, X11, X12, X15.

[0095] 5. Screen each risk source in the reachability matrix K, construct the corresponding reachable set and antecedent set, and calculate their intersection and corresponding levels. The results are shown in Table 7.

[0096] Table 7 Reachable Sets and Antecedent Sets of Each Risk Factor

[0097]

[0098]

[0099] Based on the hierarchical structure divided in the above table, the preliminary hierarchical structure model constructed in Table 1 can be updated to eliminate redundant influence relationships, and at the same time, sort out the mutual relationships among the 15 risk sources in the underwater pipeline leakage accident.

[0100] 6. Calculate the priority D of each risk factor i , and perform normalization to obtain the weight vector of each risk source, as shown in Table 8

[0101] Table 8 Weights of Each Risk Factor

[0102]

[0103] 7. Invite experts to conduct severity evaluations on all risk factors. To ensure the reliability of the questionnaire results, the survey scope was expanded. A total of 30 questionnaires were distributed. Respondents evaluated the severity of each risk source according to the alternative set V = {1, 2, 3, 4, 5} = {negligible, acceptable, controllable, critical, unacceptable}. The questionnaire statistics are shown in the following table

[0104] Table 9 Questionnaire Statistics Results

[0105]

[0106]

[0107] Perform normalization on the statistical situation of each risk source in the table under different comments to obtain the fuzzy matrix R. Calculate the severity evaluation results of each risk source, as shown in Table 10. Draw a graph of the severity weights and judgment results of the risk sources, as Figure 3 shown

[0108] Table 10 Severity Evaluation Results of Each Risk Source

[0109]

[0110] Use the weighted average method to calculate the severity of the target layer and the first-level indicators. It should be noted that Table 10 shows the overall weight values, which do not meet the normalization conditions of the first-level indicators. Therefore, it is necessary to perform normalization on the corresponding risk factor weights. The calculation results are shown in Table 11

[0111] Table 11 Importance Calculation Results

[0112]

[0113]

[0114] According to the calculation results, the severity S of the underwater pipeline leak accident (T) is 2.8391, which is between acceptable and controllable; the severity S1 of the equipment and material type risk (T1) is 2.705, which is between acceptable and controllable; the severity S2 of the operation and maintenance type risk (T2) is 2.5096, which is between acceptable and controllable; the severity S3 of the environment and external factor type risk (T3) is 3.2167, which is between controllable and critical; the severity S4 of the medium and process type risk (T4) is 2.8849, which is between acceptable and controllable.

[0115] At this point, those skilled in the art should recognize that although multiple exemplary embodiments of the present invention have been shown and described in detail herein, many other variations or modifications that conform to the principles of the present invention can still be directly determined or derived from the content disclosed in the present invention without departing from the spirit and scope of the present invention. Therefore, the scope of the present invention should be understood and determined to cover all these other variations or modifications.

Claims

1. A TrFISM-MICMAC method for analyzing the severity of underwater pipeline leakage accidents, characterized in that, The steps are as follows: Step 1: Based on the analytic hierarchy process and accident causation theory, adopt a three-level progressive model of "target layer - first-level indicators - risk sources"; starting from the target layer, decompose the risk factors of the underwater pipeline leakage accident layer by layer from top to bottom through a vertical causal chain to establish a preliminary hierarchical relationship; among them, the target layer at the highest level is the risk of underwater pipeline leakage accident, the first-level indicators in the middle layer are the macro risk categories of the direct causes of the accident, and the risk sources at the bottom layer are the minimum risk units further decomposed from the first-level indicators, that is, specific risk factors. The risk sources are mapped to the first-level indicators through logical merger and finally contribute to the target layer risk; Step 2: Invite domain experts to evaluate the influence degree between various risk factors and convert it into corresponding trapezoidal fuzzy numbers, and integrate and construct a comprehensive relationship matrix; Step 3: Construct an adjacency matrix and a reachability matrix; Step 4: Calculate the influence degree and dependence degree of each risk factor on the underwater pipeline system, and draw an influence-dependence relationship diagram of each risk factor; Step 5: Calculate the reachable set and antecedent set of each risk factor; Step 6: Calculate the priority of each risk factor for the underwater pipeline system, and after normalization, obtain the severity weight vector of each risk factor in the underwater pipeline leakage accident; Step 7: Through the fuzzy comprehensive evaluation method, quantitatively calculate the severity of the risk sources in the underwater pipeline leakage accident.

2. The TrFISM-MICMAC method according to claim 1, wherein The specific implementation process of Step 2 is as follows: Comment scale and conversion rules According to the expert evaluation results, the evaluation matrices of different experts are integrated into the initial relationship matrix by using the mean operator According to the conversion rules in the above table, it is converted into the corresponding trapezoidal fuzzy number to construct a fuzzy relationship matrix As follows: In the formula, represents the trapezoidal fuzzy number indicating the degree of mutual influence between the risk factor X i and X j ; in order to retain the uncertainty information in the trapezoidal fuzzy number as much as possible, the α-cut set is introduced; the value of α ranges from 0 to 1 and is determined by the actual operator; when α = 1, it indicates the maximum degree of uncertainty. When α = 0, it indicates the minimum degree of uncertainty; when α = 0.5, it indicates a neutral attitude; on this basis, establish an interval value relationship matrix Q as follows: where n represents the number of risk factors, denotes the element in the interval value under the α value, and respectively represent the minimum and maximum values of the α-cut set, and the calculation method is as follows: The interval value relationship matrix Q is defuzzified by the following formula to obtain the comprehensive relationship matrix Z:

3. The TrFISM-MICMAC method according to claim 2, wherein The specific implementation process of constructing the adjacency matrix and reachability matrix in Step 3 is as follows: Set the threshold θ as the mean of the comprehensive relationship matrix Z; by comparing with the threshold θ, construct the adjacency matrix L = [l ij n×n , l ij represents the influence relationship between risk factors X i and X j . The value is determined by the following formula:​ Construct the reachability matrix K as follows: Where, E represents an identity matrix with the same size as the adjacency matrix, n is the number of risk factors, and at the same time represents the actual number of iterations, k ij is an element in the reachability matrix, representing the risk factor X i and X j represent the direct or indirect reachability relationship between them.

4. The TrFISM-MICMAC method according to claim 3, wherein The specific implementation process of Step 4 is as follows: First, calculate each risk factor X t For the degree of influence and dependence on the underwater pipeline system, the calculation method is as follows: Wherein, DI t represents the influence degree of risk factor X t on the subsea manifold system, and DD t represents the dependence degree of risk factor X t on the subsea manifold system; Then, draw an influence-dependence relationship diagram of each risk factor according to the influence degree DI and dependence degree DD of each risk factor; classify all risk factors according to the four-quadrant principle, into autonomous risks (A), dependent risks (B), chain risks (C) and independent risks (D), corresponding to the third quadrant, the fourth quadrant, the first quadrant and the second quadrant respectively.

5. The TrFISM-MICMAC method according to claim 4, wherein The specific implementation process of Step 5 is as follows: First, based on the reachability matrix K, calculate the reachability set R of each risk factor through the following formula i and the antecedent set S i ; for the risk factor X i , the reachability set R i contains all risk factors reachable from X i , and the antecedent set S i contains all risk factors that can reach X i ; Then, calculate the intersection T between the two risk elements i , and determine the top-level risk factors; Then, delete the rows and columns where the risk elements at the topmost layer are located in the reachability matrix K, and calculate cyclically until all hierarchical structures and corresponding elements are determined; R i = {X i | X i ∈ X, k ji ≠ 0} S i = {X i | X i ∈ X, k ij ≠ 0} T i = R i ∩ S i Wherein, X i represents the i-th risk factor, and X represents the set of all risk factors for the underwater pipeline leak accident.

6. The TrFISM-MICMAC method according to claim 5, wherein The specific implementation process of Step 6 is as follows: Calculate the priority D of each risk factor by the following formula i , and after normalization, obtain the severity weight vector w = (w1, w2,..., w n ) T ; D i = DI i + DD i 7. The TrFISM-MICMAC method according to claim 6, wherein The specific implementation process of Step 7 is as follows: A preliminary hierarchical structure is established for the risk factors of the underwater pipeline leak accident analyzed. Each risk factor is used as the factor set C, and severity evaluation indicators are formulated according to the 5-level scale method to obtain the alternative set V; Using the weight vector w = (w1, w2, …, w n ) T obtained in Step 6 as the weight set; distributing expert questionnaires and evaluating each risk factor according to the alternative set V; Collect questionnaires and establish a fuzzy matrix R: where r iq (i = 1, 2, …, n; q = 1, 2, 3, 4, 5) represents the membership degree value of the risk factor X t under the comment q; Aggregate the weight set W and the fuzzy matrix R, and calculate the comprehensive index of the fuzzy matrix R under the influence of weights through the following formula to obtain the fuzzy evaluation result vector P; Using the weighted average method, the result vector P=(p1, p2,..., p5) is used as the weight to perform weighted calculations on each comment in the alternative set, which is used as the final severity evaluation result of the underwater pipeline leak accident.

Images