Oil and gas pipeline comprehensive risk assessment model construction method based on Vague fuzzy set and matter element extension theory

Abstract

An oil and gas pipeline comprehensive risk assessment model construction method based on a Vague fuzzy set and a matter-element extension theory belongs to the field of oil and gas chemical risk assessment, and comprises the following steps: step 1, establishing an oil and gas pipeline risk index assessment system; 2, representing an expert evaluation language by using a Vague fuzzy set, calculating a second-level risk index weight, and establishing a multi-risk factor simultaneous occurrence probability prediction model; step 3, calculating a third-level risk index weight; step 4, calculating a first-level risk index security level; and 5, verifying the accuracy of the oil and gas pipeline comprehensive risk assessment model construction method based on the Vague fuzzy set and the matter element extension theory and the strong pertinence to risk factors. Experimental results show that the method can be used for scientifically, accurately and efficiently evaluating the safety level of the risk index, and is more suitable for the field process.

Description

Technical Field

[0001] The invention belongs to the field of oil and gas chemical industry risk assessment, and in particular relates to a method for constructing an oil and gas pipeline comprehensive risk assessment model based on Vague fuzzy sets and matter-element extension theory. Background Art

[0002] Pipeline transportation is the main mode of transportation for oil and natural gas. Regular risk assessment of oil and gas pipeline systems and the implementation of risk management are of great significance for the safe operation of pipelines. Previous assessment methods were mostly based on the identification of pipeline damage factors, using semi-quantitative risk assessment methods to quantitatively analyze the causes and consequences of oil pipeline failures, and conducting current risk assessments. However, they failed to take into account the simultaneous occurrence of multiple risk factors, and risk intervention measures could not work on the risk assessment results. At the same time, risk indicators were not effectively classified and associated, and risk assessments could not fit the on-site process, and accurate assessments of the safety levels of risk indicators could not be achieved. Therefore, a scientific and efficient risk assessment model is needed to accurately assess it. Summary of the invention

[0003] In view of the above problems, the purpose of the present invention is to propose a method for constructing a comprehensive risk assessment model for oil and gas pipelines based on Vague fuzzy sets and matter-element extension theory, so as to solve the problems in the prior art that risk assessment does not fit the on-site process, risk intervention measures cannot take effect according to the risk assessment results, and risk indicators are not effectively classified and correlated, resulting in inefficient and inaccurate risk assessment models.

[0004] In order to achieve the purpose of the present invention, the present invention is implemented by the following technical solutions:

[0005] A method for constructing a comprehensive risk assessment model for oil and gas pipelines based on Vague fuzzy sets and matter-element extension theory, the method comprising the following steps:

[0006] Step 1: Establish a risk indicator system. Based on the analysis of historical accident cases of oil and gas pipelines, divide the risk indicator system into three levels. According to the application engineering direction, invite n experts to conduct language evaluation on the three-level risk indicators based on the risk indicator system. After n experts have evaluated all the risk indicators in the risk indicator system, a risk indicator evaluation matrix is ​​obtained.

[0007] Step 2: Use Vague fuzzy sets to represent the expert evaluation language and obtain the weight CC (E u ) ; and then calculate the weights of the secondary risk indicators and establish a probability prediction model for the simultaneous occurrence of multiple risk factors;

[0008] Step 3: Calculate the weights of the three-level risk indicators The third-level risk indicator weight is based on the second-level risk indicator weight. The risk value of the third-level risk index calculated by fuzzy reasoning method is i It is obtained by combining with the comprehensive weight method;

[0009] Step 4: Calculate the safety level of the first-level risk indicator. The safety level of the first-level risk indicator is determined based on the weight and safety level of the third-level risk indicator. The classical domain, section domain and matter-element to be evaluated of the third-level risk indicator are determined using the matter-element extension theory. The distance ρ and the single indicator correlation degree K between the risk level of the third-level risk indicator and its corresponding safety level are calculated. m , comprehensive index correlation K s , determine the security level variable eigenvalue K′ s , and then calculate the first-level risk indicator safety level s′;

[0010] Step 5: Conduct case analysis to verify the accuracy and strong pertinence of the method for building a comprehensive risk assessment model for oil and gas pipelines based on Vague fuzzy sets and matter-element extension theory to risk factors. Based on the evaluation data of an oil and gas pipeline built in 2002 in a certain oil field, and based on the risk indicator system established based on the analysis of historical accident cases of oil and gas pipelines, conduct analysis and put forward effective suggestions for the safety supervision of oil and gas pipelines based on the evaluation results.

[0011] Further, the step 2 comprises the following steps:

[0012] Step 2.1: Calculate the weight W(E u ), similarity of evaluation opinions S, average consistency of evaluation opinions AA, relative consistency of evaluation opinions RA, and then the weight CC (E u );

[0013] Step 2.2, calculate the weight of the secondary risk index, establish the direct impact matrix according to the impact and impact degree between the risk factors, standardize the direct impact matrix to get the indirect impact matrix, and then determine the total impact matrix, and then calculate the secondary risk index weight by the initial state information matrix of the risk index judged by experts.

[0014] Step 2.3, establish a probability prediction model for the simultaneous occurrence of multiple risk factors. According to the determined probability of occurrence of risk factors, calculate the fuzzy probability score (FPS) and fuzzy failure probability (FFP) of the probability of occurrence of risk factors. After coupling, a probability prediction model for the simultaneous occurrence of multiple risk factors is obtained.

[0015] Further, the step 1 specifically includes the following steps:

[0016] Step 1.1: Conduct expert evaluation.

[0017] According to the direction of applied engineering, n experts are invited to judge the risk indicators based on the risk indicator system. The experts' judgment levels for risk indicators are set to 6 levels: excellent, excellent, good, medium, poor, and extremely poor. The corresponding Vague values ​​are: L1<0.9, 1>, L2<0.8, 0.9>, L3<0.6, 0.8>, L4<0.4, 0.6>, L5<0.2, 0.3>, L6<0, 0.2>. n experts judge the three-level risk indicators based on the risk indicator system according to the language evaluation terms;

[0018] Step 1.2: Get Vague fuzzy sets.

[0019] After obtaining the expert evaluation data set of risk assessment indicators, the attribute language evaluation term data set is converted into a Vague value matrix A = (a nm ).

[0020] Further, the step 2 specifically includes the following steps:

[0021] Step 2.1: Calculate the weight W(E u ), similarity of evaluation opinions S, average consistency of evaluation opinions AA, relative consistency of evaluation opinions RA, weight of expert opinions CC (E u );

[0022] The weight of each expert is represented as follows:

[0023]

[0024] Where: WS j and W(E u ) are the weight value and weight factor of expert j respectively.

[0025] Assume that any pair of experts u and v give the fuzzy language value and The similarity of opinions between experts u and v is represented as follows:

[0026]

[0027] The average consistency of evaluation opinions is represented as follows:

[0028]

[0029] Where: N is the number of experts.

[0030] The relative consistency of evaluation opinions is characterized as follows:

[0031]

[0032] The weight of expert opinions is represented as follows:

[0033] CC(E u )=β×W(E u )+(1-β)×RA(E u )

[0034] Where: β represents the relaxation factor of the importance of expert opinions, which is generally taken as 0.5;

[0035] Step 2.2: Calculate the secondary risk indicator weights.

[0036] Determine the impact and degree of impact between risk factors, establish a direct impact matrix, and use 0 (no impact), 1 (low impact), 2 (medium impact), 3 (high impact) and 4 (very high impact) as the criteria for judging the degree of impact between factors. Ignoring the impact of risk factors on themselves, the diagonal elements of the matrix are all 0. The representation is as follows:

[0037]

[0038] Where: Am' is an m×m matrix, and the matrix element aa ij Indicates the degree of influence of risk factor i on factor j;

[0039] Determine the indirect impact matrix. Standardize the direct impact matrix to get the indirect impact matrix, which is represented as follows:

[0040] X=s′×A′

[0041]

[0042] Determine the total impact matrix, represented as follows:

[0043] T=X(IX) -1

[0044]

[0045] Where: I is the unit matrix, r and c are the sum of all rows and columns in the total influence matrix, r i and c j are the total impact of factor i on other factors and the total impact of factor j on other factors; (r i +c i ) represents the relative importance of factor i, also known as centrality; (r i -c i ) indicates the causal relationship between factors. If (r i -c i ) is a positive value, indicating that factor i is an influencing factor of other factors. i -c i) is a negative value, indicating that factor i is affected by other factors. Determine the initial state information matrix of the risk indicator, which can be obtained by expert judgment and is characterized as follows:

[0046]

[0047] Determine the weight of the secondary indicators, which are represented as follows:

[0048]

[0049] Where: R ei is the secondary indicator risk value, W Rei is the secondary indicator weight, Q is the risk level;

[0050] Step 2.3: Establish a probability prediction model for the simultaneous occurrence of multiple risk factors.

[0051] The trapezoidal membership function is selected as the membership function of the fuzzy number in this paper because it can generate the probability of occurrence of risk factors, which is characterized as follows:

[0052]

[0053] The center of gravity defuzzification method is used to calculate the fuzzy probability score (FPS) of the probability of occurrence of risk factors, which is characterized as follows:

[0054]

[0055] Where: X' is the defuzzified output;

[0056] Determine the probability of occurrence of risk factors. Calculate the fuzzy failure probability (FFP) and normalize the probability of occurrence of risk factors, which can be expressed as follows:

[0057]

[0058] For coupling, the risk indicators in the risk indicator system established by this technical solution are all independent, so the coupling probability of multiple risk factors occurring simultaneously can be calculated through the independent event probability calculation formula, which is characterized as follows:

[0059] PP=P(x1)×P(x2)×…。

[0060] Furthermore, the step 3 is to calculate the weights of the three-level risk indicators.

[0061] The third-level risk indicator weight is obtained by combining the third-level risk indicator risk value calculated by the fuzzy reasoning method with the comprehensive weight method on the basis of the second-level risk indicator weight, and is characterized as follows:

[0062] re i =M i ×Q

[0063]

[0064] Further, the step 4 specifically includes the following steps:

[0065] Step 4.1, use the matter-element extension theory to determine the classic domain, section domain and matter-element to be evaluated of the third-level risk indicators;

[0066] The classical domain is the range of values ​​of the characteristics of things in the matter-element, which can be characterized as follows:

[0067] R i (m)=(U m , C m , Y m )

[0068] Where: i is the serial number of the thing element, m is the number of things, U m is the name of the number of things, C m is the feature set of the thing, Y m is a set of feature value ranges;

[0069] The node domain is the total range of the value of the characteristics of things in the matter-element, which is characterized as follows:

[0070] R u (i)=(U u , C u , Y u )

[0071] Where: U u is a collection of things, C u For U u The feature set, Y u The feature set C u in U u The corresponding value range under the conditions;

[0072] The matter-element to be evaluated is the evaluation score of the matter-element, which is represented as follows:

[0073] R i (i)=(N i , C i , Y i )

[0074] Where: N i For the object to be evaluated, C i N i The feature set, Y i is the value of the corresponding feature set;

[0075] Step 4.2, based on the weights and safety levels of the three-level risk indicators, calculate the correlation between the risk level of the three-level risk indicators and their corresponding safety levels, and then calculate the safety level of the first-level risk indicators;

[0076] In the matter-element extension theory, distance refers to the relative position of a point in an interval, which can be represented as follows:

[0077]

[0078] The single index correlation refers to the correlation of the i-th feature of the object to be evaluated with respect to the classification level j, which is represented as follows:

[0079] When ρ[y ij ,y iuj ]-ρ[y ij ,y imj ]≠0:

[0080]

[0081] When ρ[y ij ,y iuj ]-ρ[y ij ,y inj ]=0:

[0082] K m (c ij )=-ρ[y ij ,y imj ]-1

[0083] The comprehensive indicator correlation refers to the degree to which the evaluation object belongs to each evaluation level, which is represented as follows:

[0084] K s =∑w ij K m (c ij )

[0085] Where: s = 1, 2…, i, w ij For feature c ij The weight coefficient of

[0086] Determine the characteristic value of the safety level variable, which is characterized as follows:

[0087]

[0088] Determine the security level, characterized as follows:

[0089]

[0090] Further, the step 5 specifically includes the following steps:

[0091] Step 5.1: Language variable conversion.

[0092] Invite 5 experts in pipeline engineering and safety to evaluate the risk based on the risk indicator system: 1 first-level risk assessment indicator F, 5 second-level risk assessment indicators S i (i=1,2,…,5), 21 third-level risk assessment indicators T j (j=1,2,…21), to judge the risk indicators;

[0093] Step 5.2: Fuzzy reasoning and comprehensive weighting method.

[0094] The fuzzy reasoning method is widely used and can play the role of comprehensive data. It is often used to calculate the risk value of risk indicators in risk assessment. This paper combines it with the expert weight to obtain the weight of the corresponding risk indicator while obtaining the risk value of the risk indicator, which greatly reduces the calculation cost.

[0095] The comprehensive weight method consists of three parts: the first part is to calculate the intermediate event weight, the second part is to calculate the single weight of the risk factor, and the third part is to calculate the comprehensive weight of the risk factor. The present invention uses the comprehensive weight method to calculate the weight of the three-level risk indicator, and uses the calculated secondary indicator weight as the intermediate event weight, and uses the normalized three-level risk indicator risk value as the single weight, thereby obtaining the comprehensive weight;

[0096] The VME-MROSC model, a comprehensive risk assessment model for oil and gas pipelines based on Vague fuzzy sets and matter-element extension theory, can simultaneously calculate the risk value and weight of the secondary risk indicator with the help of fuzzy reasoning. When weighting the tertiary risk indicators, the VME-MROSC model uses the comprehensive weight method as the main framework for calculating the weight, and uses the secondary risk indicator weight as the intermediate event weight in the comprehensive weight method. The normalized tertiary indicator risk value is used as the single event weight in the comprehensive weight method. The intermediate event weight is multiplied by the single event weight to obtain the final tertiary risk indicator weight.

[0097] Step 5.3: Risk assessment results.

[0098] From the probability values ​​of the simultaneous occurrence of multiple risk factors obtained by the simultaneous occurrence probability prediction model of multiple risk factors established in step 2, it can be seen that when the oil and gas pipeline with corrosion factor S1 is affected by third-party damage S3 and free span S5, the risk of corrosion F of the oil and gas pipeline will inevitably increase;

[0099] The risk value of the third-level risk indicator re calculated by the fuzzy reasoning method in step 3 iIt can be seen that the risk of structural degradation T8 and third-party activities T9 is at a "relatively high" level, which is the highest risk level among all three-level risk indicators. Therefore, in risk management, structural degradation T8 and third-party activities t9 must be responded to and controlled in a timely manner;

[0100] Through step 4, the first-level risk indicator safety level s′=3.0081 is obtained, that is, the corrosion risk of the oil and gas pipeline belongs to the third-level relatively safe state.

[0101] Compared with the prior art, the present invention has the following beneficial effects:

[0102] 1. Based on the case analysis of oil and gas pipeline accidents, an oil and gas pipeline risk index evaluation system is established to make the risk assessment more in line with the on-site process; then, Vague fuzzy sets are used to represent the expert judgment language, and the arithmetic average (AA) method and similarity aggregation (SA) method for calculating expert weights are combined with the fuzzy reasoning method for calculating the risk value of the evaluation index, making the calculation of the risk index weight more scientific.

[0103] 2. A probability prediction model for the simultaneous occurrence of multiple risk factors was established to solve the problem that risk intervention measures could not take effect based on risk assessment results.

[0104] 3. Combining the normalized three-level risk indicator risk value calculated by the fuzzy reasoning method with the comprehensive weight method saves calculation costs.

[0105] 4. The relationship between risk indicators and safety levels is studied using the physical-element extension theory, and the correlation of comprehensive risk indicators is obtained. The VME-MROSC model is applied to the risk assessment problem of oil and gas pipelines and compared with different risk assessment methods. The experimental results show that the proposed VME-MROSC model can accurately evaluate the safety level of risk indicators and provide effective theoretical support for the risk assessment problem of oil and gas pipelines. BRIEF DESCRIPTION OF THE DRAWINGS

[0106] Figure 1 Vague value matrix diagram for risk indicator evaluation;

[0107] Figure 2 Provide a diagram of the oil and gas pipeline risk indicator system;

[0108] Figure 3 It is the change diagram of the secondary weight comparison curve;

[0109] Figure 4 This is a graph showing the change in probability values ​​of multiple risk factors occurring simultaneously;

[0110] Figure 5 It is the risk level change diagram of the third-level risk indicators;

[0111] Figure 6This is a schematic diagram of the risk level assessment results of each model;

[0112] Figure 7 Schematic diagram of the safety level assessment results of each model. DETAILED DESCRIPTION

[0113] The following is combined with Figure 1-7 The present invention is further illustrated by the following embodiments.

[0114] In order to verify the accuracy and strong pertinence of the construction method of the comprehensive risk assessment model of oil and gas pipelines based on Vague fuzzy sets and matter-element extension theory, the evaluation data of the oil and gas pipeline built in 2002 in a certain oil field is used as the basis, and the risk index system established according to the analysis of historical accident cases of oil and gas pipelines is used to carry out analysis, and the risk factors affecting the normal operation of oil and gas pipelines are analyzed according to the evaluation results. The present invention uses Vague fuzzy sets to convert language variables, calculate the safety level of oil and gas pipelines, and puts forward effective suggestions for the safety management of oil and gas pipelines according to the specific evaluation results.

[0115] A risk indicator system is established. Based on the analysis of historical accident cases of oil and gas pipelines, the risk indicator system is divided into three levels. According to the application engineering direction, n experts are invited to conduct language evaluation on the three-level risk indicators based on the risk indicator system. After n experts have evaluated all the risk indicators in the risk indicator system, the Vague value matrix of the risk indicator evaluation can be obtained as follows: Figure 1 shown.

[0116] Based on the analysis of historical oil and gas pipeline accident cases, the present invention divides the risk indicator system into three levels. Figure 2 As shown in the figure, corrosion is used as the final oil and gas pipeline safety risk assessment indicator, that is, the first-level risk indicator F; the second-level risk indicators are corrosion factors S1, structural defects S2, third-party damage S3, natural disasters S4, and free span S5; the third-level risk indicators are internal corrosion T1, external corrosion t2, corrosion prevention T3, design factors T4, construction factors T5, laying and operation factors T6, maintenance factors T7, structural degradation t8, third-party activities T9, and third-party protection measures T10. 10 、Pipeline route T 11 , line patrol frequency t 12 , Public Education 13 、Safety Regulations 14 、Earthquake T 15 、Landslide 16 , Typhoon T 17 , burial depth T 18 , Soil type T 19 , air flow T 20 , Overhang control measures T 21 .

[0117] The comprehensive risk assessment model for oil and gas pipelines based on Vague fuzzy sets and matter-element extension theory, namely the VME-MROSC model, uses fuzzy reasoning as the main weighting method. The core idea is to calculate the safety level of the first-level risk indicators and the weight of the third-level risk indicators based on the calculation of the second-level risk indicators. Therefore, when comparing weighting methods, it is only necessary to compare the weights of the second-level risk indicators calculated by each risk assessment method. The weighting methods commonly used in risk assessment are the analytic hierarchy process and the entropy weight method. The analytic hierarchy process obtains the weights of each risk indicator based on the knowledge and cognitive judgment of the decision maker, and the entropy weight method determines the weights of each risk indicator based on the relationship between the indicator data. By importing instance data into these methods, a weight comparison chart can be obtained, such as Figure 3 shown.

[0118] Depend on Figure 3 It can be seen that the indicator weights calculated by the analytic hierarchy process are extremely variable. The maximum weighted value of the secondary risk indicator third-party damage S3 is 0.5558, and the minimum weighted value of the secondary risk indicator structural defect S2 is 0.0268, which is 0.529 different from the maximum weighted value. The risk indicator weighting difference is too large, which makes the risk indicator weighting unbalanced, indicating that this method is susceptible to subjective evaluation. The entropy weight method maintains good objectivity in weighting, but the weighting of the secondary risk indicator natural disaster S4 is too low, which cannot comprehensively reflect the expert experience, and the weighting accuracy of individual indicators is difficult to guarantee. The weighted value of the VME-MROSC model is close to the intermediate weighted value of the three weighting methods, avoiding extreme situations caused by subjective preferences and improving the weighting accuracy. Although the indicator weights differ slightly, in the present invention, the secondary risk indicator is only used as an intermediate event in the comprehensive weighting method, and will not affect the subsequent calculation of the primary risk indicator safety level and the tertiary risk indicator weight.

[0119] like Figure 4 As shown in FIG. 1 , the probability value of the simultaneous occurrence of multiple risk factors is obtained by the prediction model of the simultaneous occurrence probability of multiple risk factors established in step 2. Figure 4It can be seen that the probability of two risk factors occurring at the same time is higher, half of the probability of two risk factors occurring at the same time is higher than 0.02, and the highest probability is close to 0.1; the probability of two risk factors occurring at the same time is basically between 0 and 0.02, and the probability of most two risk factors occurring at the same time is close to 0, which is consistent with the fact that the simultaneous occurrence of more than two risk factors is a rare engineering reality. There are three probability peaks in the two risk factors, namely, the simultaneous occurrence of corrosion factor S1 and third-party damage S3, the simultaneous occurrence of corrosion factor S1 and free span S5, and the simultaneous occurrence of third-party damage S3 and free span S5, which also leads to an increase in the probability of corrosion factor S1, third-party damage S3 and free span S5 occurring at the same time. This shows that when the oil and gas pipeline with corrosion factors is affected by third-party damage and free span, the risk of corrosion of the oil and gas pipeline will inevitably increase. Therefore, in risk management, targeted measures need to be taken to avoid the occurrence of the above situation.

[0120] like Figure 5 As shown in the figure, the risk value of the third-level risk indicator re calculated by the fuzzy reasoning method in step 3 i .Depend on Figure 5 It can be seen that the risk of structural degradation T8 and third-party activities T9 is at a "relatively high" level, which is the highest risk level among all three-level risk indicators. Therefore, in risk management, structural degradation T8 and third-party activities T9 should be promptly responded to and controlled. The same data is then brought into the AHP-TOPSIS risk assessment model, the risk assessment model based on the structural entropy weight method, and the risk assessment model based on the fuzzy reasoning method for comparative study. These models are selected because they are widely used to solve risk assessment problems. The respective risk level assessment results are shown in Figure 6 As shown in the above risk level assessment results, it can be seen that in the above three risk assessment models, structural degradation T8 and third-party activities T9 have the highest risk levels in their respective assessment results, while corrosion prevention T3 and safety regulations T 14 、Earthquake T 15 The risk level is the lowest in their respective assessment results. The results show that the VME-MROSC model has a high degree of consistency with the other three risk assessment models. The risk level assessment results obtained by the risk assessment model based on the structural entropy weight method are too high, making the assessment results inaccurate; the risk level assessment results obtained by the risk assessment model based on the fuzzy reasoning method are too concentrated, resulting in the inability to accurately distinguish the risk levels of different risk indicators. In addition, compared with the above three risk assessment models, the VME-MROSC model is more accurate in identifying risk indicators at the same risk level, with corrosion prevention T3, safety regulations T4, and others. 14 、Earthquake T 15 and landslide T 16 , Typhoon T 17 , burial depth T18 , Soil type T 19 For example, the VME-MROSC model, the risk assessment model based on the structural entropy weight method, and the risk assessment model based on the fuzzy reasoning method all take corrosion prevention T3, safety regulations T 14 、Earthquake T 15 The VME-MROSC model, the improved AHP-TOPSIS model and the risk assessment model based on fuzzy reasoning method all classified the landslide T 16 , Typhoon T 17 , burial depth T 18 , Soil type T 19 The above analysis shows that the VME-MROSC model is more accurate in assessing the risk level of risk indicators.

[0121] Through step 4, the first-level risk indicator safety level s′=3.0081 is obtained, that is, the corrosion risk of the oil and gas pipeline belongs to the third level of relatively safe state. And the same data is brought into a variety of different risk assessment models to verify the evaluation results. The selected models include risk assessment models based on cloud models, risk assessment models based on hierarchical analysis method, risk assessment models based on grey correlation method, risk assessment models based on fuzzy reasoning method, risk assessment models based on cooperative game method, and risk assessment models based on comprehensive weight method. These models were selected because they have been used to solve the risk assessment problem of oil and gas pipelines. The respective safety level assessment results s′ were obtained: 3.0480, 2.8393, 3.1623, 3.1847, 3.0990, 3.1603. The safety level assessment results of oil and gas pipelines are as follows Figure 7 As shown. Figure 7 It can be seen that the safety level assessment results s′ of the above six methods are all close to the third level of relatively safe state, which is consistent with the safety level assessment results obtained by the VME-MROSC model, indicating that the VME-MROSC model can produce reliable assessment results.

[0122] Oil and gas pipeline risk assessment technology is an important way to ensure that oil and gas pipelines can work safely. Aiming at the problems of uncertainty in the expert's judgment of risk indicators, weak pertinence of risk response measures and inaccurate assessment results in the process of oil and gas pipeline risk assessment, the present invention proposes a method for constructing an oil and gas pipeline comprehensive risk assessment model based on Vague fuzzy sets and matter-element extension theory, combines fuzzy reasoning method and comprehensive weight method, and establishes a probability prediction model for the simultaneous occurrence of multiple risk factors, thereby improving the accuracy of oil and gas pipeline risk assessment and the pertinence of measures for risk factors. The VME-MROSC model uses Vague fuzzy sets to effectively overcome the influence of expert judgment hesitation. By combining the judgment information of five experts on risk indicators and calculating the consistency of the judgment, the uncertainty of risk assessment is effectively reduced. A probability prediction model for the simultaneous occurrence of multiple risk factors is established, and then several risk factors that need special attention are obtained, thereby enhancing the pertinence of risk factor response measures. Combining fuzzy reasoning method with comprehensive weight method makes the obtained risk indicator weight more reliable and reduces the calculation cost, thereby improving the accuracy of the assessment results. Finally, the VME-MROSC model is applied to the risk assessment of oil and gas pipelines and compared with three risk assessment models. The assessment results of the VME-MROSC model are more accurate.

[0123] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Other modifications or equivalent substitutions made to the technical solution of the present invention by ordinary technicians in the field should be included in the scope of the claims of the present invention as long as they do not depart from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for constructing a comprehensive risk assessment model for oil and gas pipelines based on Vague fuzzy sets and matter-element extension theory, characterized by: The construction method comprises the following steps: Step 1: Establish a risk indicator system. Based on the analysis of historical accident cases of oil and gas pipelines, divide the risk indicator system into three levels. According to the application engineering direction, invite b experts to conduct language evaluation on the three-level risk indicators based on the risk indicator system. After n experts have evaluated all the risk indicators in the risk indicator system, a risk indicator evaluation matrix is ​​obtained. Step 2: Use Vague fuzzy sets to represent the expert evaluation language and obtain the weight CC (E u ) ; and then calculate the weights of the secondary risk indicators and establish a probability prediction model for the simultaneous occurrence of multiple risk factors; Step 3: Calculate the weights of the three-level risk indicators The third-level risk indicator weight is based on the second-level risk indicator weight. The risk value of the third-level risk index calculated by fuzzy reasoning method is i It is obtained by combining with the comprehensive weight method; Step 4: Calculate the safety level of the first-level risk indicator. The safety level of the first-level risk indicator is determined based on the weight and safety level of the third-level risk indicator. The classical domain, section domain and matter-element to be evaluated of the third-level risk indicator are determined using the matter-element extension theory. The distance ρ and the single indicator correlation degree K between the risk level of the third-level risk indicator and its corresponding safety level are calculated. m , comprehensive index correlation K s , determine the safety level variable characteristic value K' s , and then calculate the first-level risk indicator safety level s'; Step 5: Conduct case analysis to verify the accuracy and strong pertinence of the method for building a comprehensive risk assessment model for oil and gas pipelines based on Vague fuzzy sets and matter-element extension theory to risk factors. Based on the evaluation data of an oil and gas pipeline built in 2002 in a certain oil field, and based on the risk indicator system established based on the analysis of historical accident cases of oil and gas pipelines, conduct analysis and put forward effective suggestions for the safety supervision of oil and gas pipelines based on the evaluation results.

2. The method for constructing a comprehensive risk assessment model for oil and gas pipelines based on Vague fuzzy sets and matter-element extension theory according to claim 1 is characterized in that: The step 2 comprises the following steps: Step 2.1: Calculate the weight W(E u ), similarity of evaluation opinions S, average consistency of evaluation opinions AA, relative consistency of evaluation opinions RA, and then the weight CC (E u ); Step 2.2, calculate the weight of the secondary risk index, establish the direct impact matrix according to the impact and impact degree between the risk factors, standardize the direct impact matrix to get the indirect impact matrix, and then determine the total impact matrix, and then calculate the secondary risk index weight by the initial state information matrix of the risk index judged by experts. Step 2.3, establish a probability prediction model for the simultaneous occurrence of multiple risk factors. According to the determined probability of occurrence of risk factors, calculate the fuzzy probability score (FPS) and fuzzy failure probability (FFP) of the probability of occurrence of risk factors. After coupling, a probability prediction model for the simultaneous occurrence of multiple risk factors is obtained.

3. The method for constructing a comprehensive risk assessment model for oil and gas pipelines based on Vague fuzzy sets and matter-element extension theory according to claim 1 is characterized by: The step 1 specifically comprises the following steps: Step 1.1: Conduct expert evaluation. According to the application engineering direction, n experts are invited to judge the risk indicators based on the risk indicator system. The experts' judgment levels for risk indicators are set to 6 levels: excellent, excellent, good, medium, poor, and extremely poor. The corresponding Vague values ​​are: L1<0.9,1>, L2<0.8,0.9>, :3<0.6,0.8>, L4<0.4,0.6>, L5<0.2,0.3>, L6M0,0.2<. n experts judge the three-level risk indicators based on the language evaluation terms and the risk indicator system; Step 1.2: Get Vague fuzzy sets. After obtaining the expert evaluation data set of risk assessment indicators, the attribute language evaluation term data set is converted into a Vague value matrix A = (a nm ).

4. The method for constructing a comprehensive risk assessment model for oil and gas pipelines based on Vague fuzzy sets and matter-element extension theory according to claim 2 is characterized by: The step 2 specifically includes the following steps: Step 2.1: Calculate the weight W(E u ), similarity of evaluation opinions S, average consistency of evaluation opinions AA, relative consistency of evaluation opinions RA, weight of expert opinions CC (E u ); The weight of each expert is represented as follows: Where: WS j and W(E u ) are the weight value and weight factor of expert j respectively; Assume that any pair of experts u and v give the fuzzy language value and The similarity of opinions between experts u and v is represented as follows: The average consistency of evaluation opinions is represented as follows: Where: N is the number of experts, The relative consistency of the evaluation opinions is characterized as follows: The weight of expert opinions is represented as follows: CC(E u )=β×W(E u )+(1-β)×RA(E u ) Where: β represents the relaxation factor of the importance of expert opinions, which is generally taken as 0.5; Step 2.2: Calculate the secondary risk indicator weights. Determine the impact and degree of impact between risk factors, establish a direct impact matrix, and use 0 (no impact), 1 (low impact), 2 (medium impact), 3 (high impact) and 4 (very high impact) as the criteria for judging the degree of impact between factors. Ignoring the impact of the risk factor on itself, the diagonal elements of the matrix are all 0, as represented as follows: Where: A′ is an m×m matrix, and the matrix elements aa ij Indicates the degree of influence of risk factor i on factor j; Determine the indirect impact matrix. Standardize the direct impact matrix to get the indirect impact matrix, which is represented as follows: X=s′×A′ Determine the total impact matrix, represented as follows: T=X(I-X)- 1 Where: I is the unit matrix, r and c are the sum of all rows and columns in the total influence matrix, r i and c j are the total impact of factor i on other factors and the total impact of factor j on other factors, (r i +c i ) represents the relative importance of factor i, also known as centrality; (r i -c i ) indicates the causal relationship between factors. If (r i -c i ) is a positive value, indicating that factor i is an influencing factor of other factors. i -c i ) is a negative value, indicating that factor i is affected by other factors. Determine the initial state information matrix of the risk indicator, which can be obtained by expert judgment and is characterized as follows: Determine the weight of the secondary indicators, which are represented as follows: Where: R ei is the secondary indicator risk value, W Rei is the secondary indicator weight, Q is the risk level; Step 2.3: Establish a probability prediction model for the simultaneous occurrence of multiple risk factors. The trapezoidal membership function is selected as the membership function of the fuzzy number in this paper because it can generate the probability of occurrence of risk factors, which is characterized as follows: The center of gravity defuzzification method is used to calculate the fuzzy probability score (FPS) of the probability of occurrence of risk factors, which is characterized as follows: Where: X' is the defuzzified output; Determine the probability of occurrence of risk factors, calculate the fuzzy failure probability (FFP), and normalize the probability of occurrence of risk factors, which is represented as follows: The risk indicators in the risk indicator system established by this technical solution are all independent. The coupling probability of multiple risk factors occurring simultaneously can be calculated through the independent event probability calculation formula, which is characterized as follows: PP=P(x1)×P(x2)×…。 5. The method for constructing a comprehensive risk assessment model for oil and gas pipelines based on Vague fuzzy sets and matter-element extension theory according to claim 1 is characterized by: The step 3 is to calculate the third-level risk indicator weight. The third-level risk indicator weight is obtained by combining the third-level risk indicator risk value calculated by the fuzzy reasoning method with the comprehensive weight method on the basis of the second-level risk indicator weight, and is characterized as follows: re i =M i ×Q 6. The method for constructing a comprehensive risk assessment model for oil and gas pipelines based on Vague fuzzy sets and matter-element extension theory according to claim 1 is characterized by: The step 4 specifically comprises the following steps: Step 4.1: Use the matter-element extension theory to determine the classic domain, section domain and matter-element to be evaluated of the three-level risk indicators. The classical domain is the range of values ​​of the characteristics of things in the matter-element, which can be characterized as follows: R i (m)=(U m ,C m ,Y m ) Where: i is the serial number of the thing element, m is the number of things, U m is the name of the number of things, C m is the feature set of the thing, Y m is a set of feature value ranges; The node domain is the total range of the value of the characteristics of things in the matter-element, which is characterized as follows: R u (i)=(U u ,C u ,Y u ) Where: U u is a collection of things, C u For U u The feature set, Y u The feature set C u in U u The corresponding value range under the conditions; The matter-element to be evaluated is the evaluation score of the matter-element, which is represented as follows: R i (i)=(N i ,C i ,Y i ) Where: N i For the object to be evaluated, C i N i The feature set, Y i is the value of the corresponding feature set; Step 4.2, based on the weights and safety levels of the three-level risk indicators, calculate the correlation between the risk level of the three-level risk indicators and their corresponding safety levels, and then calculate the safety level of the first-level risk indicators; In the matter-element extension theory, distance refers to the relative position of a point in an interval, which can be represented as follows: The single index correlation refers to the correlation of the i-th feature of the object to be evaluated with respect to the classification level j, which is represented as follows: When ρ[y ij ,y iuj -ρ[y ij ,y imj ≠0: When ρ[y ij , y iuj - ρ[y ij , y imj = 0: K m (c ij )=-ρ[y ij ,y imj ]-1 The comprehensive indicator correlation refers to the degree to which the evaluation object belongs to each evaluation level, which is represented as follows: K s =∑w ij K m (c ij ) Where: s = 1, 2…, i, w ij For feature c ij The weight coefficient of Determine the characteristic value of the safety level variable, which is characterized as follows: Determine the security level, characterized as follows:

7. The method for constructing a comprehensive risk assessment model for oil and gas pipelines based on Vague fuzzy sets and matter-element extension theory according to claim 1 is characterized by: The step 5 specifically comprises the following steps: Step 5.1: Perform language variable conversion. Invite 5 experts in pipeline engineering and safety to evaluate the risk based on the risk indicator system: 1 first-level risk assessment indicator F, 5 second-level risk assessment indicators S i (i=1,2,…,5), 21 third-level risk assessment indicators T j (j=1,2,…21), to judge the risk indicators; Step 5.2: Fuzzy reasoning and comprehensive weighting method. The fuzzy reasoning method is often used to calculate the risk value of risk indicators in risk assessment. This technical solution combines it with the expert weight to obtain the weight of the corresponding risk indicator while obtaining the risk value of the risk indicator; The comprehensive weight method consists of three parts: the first part is to calculate the intermediate event weight, the second part is to calculate the single weight of the risk factor, and the third part is to calculate the comprehensive weight of the risk factor. The present invention uses the comprehensive weight method to calculate the weight of the three-level risk indicator, and uses the calculated secondary indicator weight as the intermediate event weight, and uses the normalized three-level risk indicator risk value as the single weight, thereby obtaining the comprehensive weight; The VME-MROSC model, a comprehensive risk assessment model for oil and gas pipelines based on Vague fuzzy sets and matter-element extension theory, can simultaneously calculate the risk value and weight of the secondary risk indicator with the help of fuzzy reasoning. When weighting the tertiary risk indicators, the VME-MROSC model uses the comprehensive weight method as the main framework for calculating the weight, and uses the secondary risk indicator weight as the intermediate event weight in the comprehensive weight method. The normalized tertiary indicator risk value is used as the single event weight in the comprehensive weight method. The intermediate event weight is multiplied by the single event weight to obtain the final tertiary risk indicator weight. Step 5.3: Obtain risk assessment results. From the probability values ​​of the simultaneous occurrence of multiple risk factors obtained by the simultaneous occurrence probability prediction model of multiple risk factors established in step 2, it can be seen that when the oil and gas pipeline with corrosion factor S1 is affected by third-party damage S3 and free span S5, the risk of corrosion F of the oil and gas pipeline will inevitably increase; The risk value of the third-level risk indicator re calculated by the fuzzy reasoning method in step 3 i It can be seen that the risk of structural degradation T8 and third-party activities T9 is at a "relatively high" level, which is the highest risk level among all three-level risk indicators. Therefore, in risk management, structural degradation T8 and third-party activities T9 must be responded to and controlled in a timely manner; Through step 4, the first-level risk indicator safety level s′=3.0081 is obtained, that is, the corrosion risk of the oil and gas pipeline belongs to the third-level relatively safe state.

Images