Refinery enterprise water pipeline risk evaluation method based on Bayesian network

Abstract

The invention provides a refinery enterprise water pipeline risk evaluation method based on a Bayesian network, and belongs to the technical field of pipeline risk evaluation. According to the technical scheme, the refinery enterprise water pipeline risk evaluation method based on the Bayesian network is characterized by comprising the specific steps that previous pipeline failure cases are collected, and a historical pipeline failure database is constructed; identifying failure factors through historical data in the historical failure database, and establishing an accident tree based on refinery enterprise water pipeline failure; determining the prior failure probability of the failure factors, and constructing a Bayesian network structure; calculating the fuzzy probability of the root node, and determining a correction factor; and calculating the posterior failure probability of the refinery enterprise water pipeline. The refinery enterprise water pipeline risk evaluation method based on the Bayesian network has the beneficial effects that the subjectivity of water pipeline risk factor identification is reduced, and meanwhile, the accuracy and effectiveness of a risk prediction evaluation result are improved.

Description

Technical Field

[0001] The invention belongs to the technical field of pipeline risk assessment, and in particular relates to a Bayesian network-based water pipeline risk assessment method for a refinery. Background Art

[0002] As the service life of water supply pipelines increases, pipeline corrosion and perforation problems become increasingly prominent, which not only causes waste of water resources and energy, but also seriously affects the stability of water supply. The traditional water supply pipeline management model currently adopted by refining and chemical enterprises is time-consuming and labor-intensive and has a lot of blindness. It can no longer meet the needs of on-site generation and urgently needs to be improved. The traditional management model of the management department to eliminate hidden dangers point by point can only be "twice the effort for half the result" in solving the problem of water supply pipelines, and it cannot support the decision-making level to make overall planning and systematically solve the problem of water supply pipelines.

[0003] my country's cities are in a period of rapid development, and the status of pipelines has changed with time and the expansion of the pipeline network. Therefore, adopting a reasonable method system to evaluate pipelines is a far-reaching and challenging task. At present, relevant standard methods at home and abroad are mostly aimed at dangerous medium transportation pipelines in oil and gas stations or chemical plants, but not enough attention is paid to water pipelines. How to better reflect the characteristics of water pipelines and provide accurate risk assessment information is the key and difficult content that needs to be solved in the process of developing water pipeline risk assessment methods.

[0004] In view of the above problems, the present invention proposes a dynamic evaluation method based on the changes and correlation of measured data, which is of great significance to the risk management of water pipelines in refining enterprises. Summary of the invention

[0005] The purpose of the present invention is to provide a Bayesian network-based water pipeline risk assessment method for a refinery to reduce the subjectivity of risk factor identification for water pipelines and improve the accuracy and effectiveness of risk prediction and assessment results.

[0006] The present invention is achieved through the following measures: a method for risk assessment of water pipelines in refineries based on Bayesian networks, characterized in that the specific steps are:

[0007] S1. Collect previous pipeline failure cases and build a historical pipeline failure database;

[0008] S2. Identify failure factors through historical data in the historical failure database and establish an accident tree based on the failure of the water pipeline of the refinery;

[0009] S3, determine the prior failure probability of the failure factor and construct a Bayesian network structure;

[0010] S4, calculating the fuzzy probability of the root node and determining the correction factor;

[0011] S5. Calculate the a posteriori failure probability of the water pipelines used by the refinery.

[0012] Step S2 is specifically as follows:

[0013] S21. Analyze the failure factors of pipeline failure data to determine the basic events and intermediate events that lead to failure;

[0014] S22. Construct an accident tree according to the relationship between the basic events and the intermediate events.

[0015] Step S3 is specifically as follows:

[0016] S31. Determine the prior failure probability of each failure factor based on existing pipeline failure cases;

[0017] S32, determining the relationship between the parent node and the child node according to the prior failure probability and the established fault tree, and converting the network structure of the fault tree into a Bayesian network structure;

[0018] S33. Determine the fuzzy subsets of the root node and intermediate node failure probabilities in the Bayesian network.

[0019] Step S4 is specifically as follows:

[0020] S41. Consider the pipeline, collect the distribution information of the above parameters, and prepare a failure factor classification table, which is divided into primary, secondary and tertiary levels; the higher the level, the greater the possibility of causing pipeline failure;

[0021] S42, assigning values ​​to the parameters in the previous step according to the structural importance, preparing a failure factor weight table, wherein the failure factor weight table sets the weight of each factor according to expert experience and historical data;

[0022] S43. Convert each level into a corresponding fuzzy number; if it is divided into three levels, "level one" corresponds to "0.3", "level two" corresponds to "0.5", and "level three" corresponds to "0.7"; if it is divided into two levels, "level one" corresponds to "0.3" and "level two" corresponds to "0.7".

[0023] S44, converting the fuzzy number into a fuzzy probability through a formula; the calculation formula of the fuzzy probability is as follows:

[0024]

[0025]

[0026]

[0027] Among them, FP is fuzzy probability, FPS is fuzzy number, r is average failure probability; the average failure probability of the pipeline system is determined according to the statistical data of the failure database; K and k are the intermediate parameter values ​​of the calculation process;

[0028] S45, setting the correction factor of level three (or level two) to "1", and the correction factors corresponding to other levels are the ratios of the fuzzy probabilities corresponding to the respective levels to the fuzzy probabilities corresponding to level three (or level two);

[0029] Step S5 is specifically as follows: the failure probability calculation model adopts the following formula:

[0030]

[0031] Among them, P i is the pipeline failure probability caused by failure factor i, R i is the prior failure probability corresponding to failure factor i, F i is the failure probability correction factor corresponding to failure factor i.

[0032] The specific steps also include:

[0033] S6. Calculate the failure risk of water pipelines based on the a posteriori failure probability and establish a classification table for the consequences of pipeline failure.

[0034] The step S6 is specifically as follows:

[0035] S61. Develop risk assessment standards for various failure factors of water pipelines in refineries;

[0036] S62. Establish a pipeline failure consequence classification table and failure consequence scoring criteria;

[0037] S63. Calculate the risk of water pipelines in refineries;

[0038] S64. Establish a classification table for the likelihood of accidents occurring in water pipelines of refineries and petrochemical enterprises and a classification table for risk index.

[0039] When analyzing the failure factors of pipeline failure data, a comprehensive failure factor analysis is conducted from three aspects: pipeline corrosion, misoperation, and third-party damage. Specifically, it includes pipe age, pipe material, properties of the conveying medium, anti-corrosion layer conditions, pipeline burial depth, pipe diameter, pipeline penetration and span conditions, pipeline laying conditions, and pipeline leakage conditions.

[0040] Compared with the prior art, the beneficial effects of the present invention are as follows: based on the idea of ​​combining subjective and objective data, a Bayesian network-based risk assessment method for water pipelines in refineries is proposed. The method obtains the main influencing factors of water pipeline failure through the failure tree of water pipelines in refineries, classifies each factor and performs fuzzy processing to obtain the fuzzy correction factor of each factor level, sets the weight of each factor according to expert experience and historical data, calculates the failure probability of water pipelines in refineries through Bayesian networks, and calculates the pipeline risk value and classifies the risk level in combination with the failure consequence grading evaluation table. This method can reduce the subjectivity of risk factor identification, improve the accuracy of risk prediction, and the failure probability can be updated in real time, thereby providing a more reasonable and effective basis for risk assessment of water pipelines in refineries. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A schematic diagram of a process of a method for risk assessment of water pipelines for a refinery based on a Bayesian network in an embodiment of the present invention;

[0042] Figure 2 This is an accident tree diagram of a water pipeline failure in a refinery in an embodiment of the present invention;

[0043] Figure 3 Schematic diagram of the Bayesian network structure of the failure probability of the water pipeline of the refinery in the embodiment of the present invention. DETAILED DESCRIPTION

[0044] In order to clearly illustrate the technical features of this solution, this solution is described below through a specific implementation method.

[0045] Embodiment 1:

[0046] See also Figure 1-Figure 3 , a method for risk assessment of water pipelines in refineries based on Bayesian networks, characterized in that the specific steps are:

[0047] S1. Collect previous pipeline failure cases and build a historical pipeline failure database;

[0048] S2. Identify failure factors through historical data in the historical failure database and establish an accident tree based on the failure of the water pipeline of the refinery;

[0049] S3, determine the prior failure probability of the failure factor and construct a Bayesian network structure;

[0050] S4, calculating the fuzzy probability of the root node and determining the correction factor;

[0051] S5. Calculate the a posteriori failure probability of the water pipelines used by the refinery.

[0052] Step S2 is specifically as follows:

[0053] S21. Analyze the failure factors of pipeline failure data to determine the basic events and intermediate events that lead to failure;

[0054] S22. Construct an accident tree according to the relationship between the basic events and the intermediate events.

[0055] Step S3 is specifically as follows:

[0056] S31. Determine the prior failure probability of each failure factor based on existing pipeline failure cases;

[0057] S32, determining the relationship between the parent node and the child node according to the prior failure probability and the established fault tree, and converting the network structure of the fault tree into a Bayesian network structure;

[0058] S33. Determine the fuzzy subsets of the root node and intermediate node failure probabilities in the Bayesian network.

[0059] Step S4 is specifically as follows:

[0060] S41. Consider the pipeline, collect the distribution information of the above parameters, and prepare a failure factor classification table, which is divided into primary, secondary and tertiary levels; the higher the level, the greater the possibility of causing pipeline failure;

[0061] S42, assigning values ​​to the parameters in the previous step according to the structural importance, preparing a failure factor weight table, wherein the failure factor weight table sets the weight of each factor according to expert experience and historical data;

[0062] S43. Data processing: Convert each level into a corresponding fuzzy number; if it is divided into three levels, "level one" corresponds to "0.3", "level two" corresponds to "0.5", and "level three" corresponds to "0.7"; if it is divided into two levels, "level one" corresponds to "0.3" and "level two" corresponds to "0.7".

[0063] S44, converting the fuzzy number into a fuzzy probability through a formula; the calculation formula of the fuzzy probability is as follows:

[0064]

[0065]

[0066]

[0067] Among them, FP is fuzzy probability, FPS is fuzzy number, r is average failure probability; the average failure probability of the pipeline system is determined according to the statistical data of the failure database; K and k are the intermediate parameter values ​​of the calculation process;

[0068] S45, setting the correction factor of level three (or level two) to "1", and the correction factors corresponding to other levels are the ratios of the fuzzy probabilities corresponding to the respective levels to the fuzzy probabilities corresponding to level three (or level two);

[0069] Step S5 is specifically as follows: the failure probability calculation model adopts the following formula:

[0070]

[0071] Among them, P i is the pipeline failure probability caused by failure factor i, R i is the prior failure probability corresponding to failure factor i, F i is the failure probability correction factor corresponding to failure factor i.

[0072] The specific steps also include:

[0073] S6. Calculate the failure risk of water pipelines based on the a posteriori failure probability and establish a classification table for the consequences of pipeline failure.

[0074] The step S6 is specifically as follows:

[0075] S61. Develop risk assessment standards for various failure factors of water pipelines in refineries;

[0076] S62. Establish a pipeline failure consequence classification table and failure consequence scoring criteria;

[0077] S63. Calculate the risk of water pipelines in refineries;

[0078] S64. Establish a classification table for the likelihood of accidents occurring in water pipelines of refineries and petrochemical enterprises and a classification table for risk index.

[0079] When analyzing the failure factors of pipeline failure data, a comprehensive failure factor analysis is conducted from three aspects: pipeline corrosion, misoperation, and third-party damage. Specifically, it includes pipe age, pipe material, properties of the conveying medium, anti-corrosion layer conditions, pipeline burial depth, pipe diameter, pipeline penetration and crossing conditions, pipeline laying conditions, and pipeline accidents (leakage conditions).

[0080] Embodiment 2:

[0081] See also Figure 1-Figure 3 , a risk assessment method for water pipelines in refining and petrochemical enterprises based on Bayesian networks, the specific steps are:

[0082] S1. Collect previous pipeline failure cases and build a historical pipeline failure database;

[0083] S2. Identify failure factors through historical data in the historical failure database and establish an accident tree based on the failure of the water pipeline of the refinery;

[0084] S3, determine the prior failure probability of the failure factor and construct a Bayesian network structure;

[0085] S4, calculating the fuzzy probability of the root node and determining the correction factor;

[0086] S5. Calculate the a posteriori failure probability of the water pipelines used by the refinery.

[0087] Step S2 is specifically as follows:

[0088] S21. By analyzing the failure factors of pipeline failure data, the basic events and intermediate events that lead to failure are determined. The factors affecting the failure of water pipelines in refineries are complex, and most accidents are caused by multiple factors acting simultaneously. By analyzing and studying the failure record data of water pipelines in refineries, a comprehensive failure factor analysis is conducted from three aspects: pipeline corrosion, misoperation, and third-party damage, including pipe age, pipe material, properties of the conveying medium, anti-corrosion layer, pipeline burial depth, pipe diameter, pipeline crossing, pipeline laying, and pipeline leakage. The basic events of failure accidents are determined using the accident tree method, such as Figure 1 As shown in the figure, the fault tree is as follows Figure 2 As shown, it can be divided into 4 intermediate nodes and 14 root nodes.

[0089] S22. Construct an accident tree according to the relationship between the basic events and the intermediate events.

[0090] Step S3 is specifically as follows:

[0091] S31. Determine the prior failure probability of each failure factor based on existing pipeline failure cases;

[0092] S32, determining the relationship between the parent node and the child node according to the prior failure probability and the established fault tree, and converting the network structure of the fault tree into a Bayesian network structure;

[0093] S33. Determine the fuzzy subsets of the root node and intermediate node failure probabilities in the Bayesian network.

[0094] Step S4 is specifically as follows:

[0095] S41. Consider the pipeline, collect the distribution information of the above parameters, and prepare a failure factor classification table, which is divided into primary, secondary and tertiary levels; the higher the level, the greater the possibility of causing pipeline failure;

[0096] In terms of pipe age, pipelines that have been in use for 1-20 years are classified as Level 1, 21-40 years as Level 2, and 41-50 years and above as Level 3.

[0097] For pipe diameter, pipe diameter <200mm is grade one, pipe diameter between 200-400mm is grade two (including 200mm, excluding 400mm), and pipe diameter ≥400mm is grade three;

[0098] For pipeline laying, underground laying is classified as level one, partially underground laying is classified as level two, and above-ground laying is classified as level three;

[0099] For pipeline burial depth, if the buried depth is more than 2m, it is level one, and if it is less than 2m, it is level two.

[0100] For pipe materials, FRP is grade one, 304 stainless steel is grade two, and 20# carbon steel is grade three;

[0101] For the transport medium, circulating water, production water supply, domestic water supply, production wastewater, domestic sewage, deionized water and rainwater are classified as level one, and fire water is classified as level two;

[0102] For pipeline crossings, the situation without crossings is level one, and the situation with crossings is level two.

[0103] Anti-corrosion layer condition: obtain the test data of the water pipe anti-corrosion layer. Qualified is level one; no data is level two; unqualified is level three;

[0104] For leakage frequency, no leakage is level one, occasional leakage is level two, and frequent leakage is level three.

[0105] S42. Assign values ​​to the parameters in the previous step according to the structural importance, and prepare a failure factor weight table. The failure factor weight table sets the weight of each factor according to expert experience and historical data, specifically:

[0106] parameter Age Pipe diameter Laying situation Buried Depth Pipe Conveying medium Wear across Anti-corrosion Leakage frequency Weight 0.20 0.05 0.06 0.06 0.04 0.19 0.10 0.10 0.20

[0107] S43. Convert each level into a corresponding fuzzy number; if it is divided into three levels, "level one" corresponds to "0.3", "level two" corresponds to "0.5", and "level three" corresponds to "0.7"; if it is divided into two levels, "level one" corresponds to "0.3" and "level two" corresponds to "0.7".

[0108] S44, converting the fuzzy number into a fuzzy probability through a formula; the calculation formula of the fuzzy probability is as follows:

[0109]

[0110]

[0111]

[0112] Among them, FP is fuzzy probability, FPS is fuzzy number, r is average failure probability; the average failure probability of the pipeline system is determined according to the statistical data of the failure database; K and k are the intermediate parameter values ​​of the calculation process;

[0113] S45, setting the correction factor of level three (or level two) to "1", and the correction factors corresponding to other levels are the ratios of the fuzzy probabilities corresponding to the respective levels to the fuzzy probabilities corresponding to level three (or level two);

[0114] Step S5 is specifically as follows: the failure probability calculation model adopts the following formula:

[0115]

[0116] Among them, P i is the pipeline failure probability caused by failure factor i, R i is the prior failure probability corresponding to failure factor i, F i is the failure probability correction factor corresponding to failure factor i.

[0117] The specific steps also include:

[0118] S6. Calculate the failure risk of water pipelines based on the a posteriori failure probability and establish a classification table for the consequences of pipeline failure.

[0119] The step S6 is specifically as follows:

[0120] S61. Determine the scope of impact based on the enterprise's situation and formulate risk assessment standards for various failure factors of the enterprise's water pipelines;

[0121] S62. Establish a pipeline failure consequence classification table and failure consequence scoring criteria.

[0122] Embodiment three:

[0123] Based on Example 2, the consequences caused by leakage of water pipelines in refineries are mainly divided into:

[0124] (1) Risk of personal injury. High-pressure water can directly or indirectly cause injuries to workers or property, and emergency work can easily cause electrical hazards, mechanical injuries, and falls from heights;

[0125] (2) Risk of equipment damage. When a water pipeline leaks, sewage may flow on the surface or seep underground, affecting or harming surrounding equipment, or even causing equipment damage.

[0126] Referring to the regional grades along the pipeline specified by the pipeline company, the 20m area on both sides of the water pipeline and the 10m area upstream and downstream of the leakage point are taken as the impact range of the water pipeline accident. The impact range here is defined as the boundary area. Table 1 shows the risk assessment standard for water pipeline failure. Different refineries have different scales and different geographical environments, and they need to be determined according to their specific circumstances. The scores corresponding to each consequence level of AG are 1-7 points respectively; the calculation process is the sum of the corresponding scores of the two parts of personal damage and property loss impact.

[0127]

[0128]

[0129] 1. When there are non-dangerous goods storage areas within 20m on both sides of the leak:

[0130] The affected area of ​​the leakage point is 2m 2 The value is 1 person / m2 for densely populated industrial places. 2 Calculation: if there are important equipment and buildings in the factory area, the price is 50,000 yuan / m 2 calculate.

[0131] Personal injury (major safety accident resulting in death or serious injury: 3 to 9 people seriously injured in the boundary area; 1 to 2 people seriously injured outside the boundary area). Property loss (direct economic loss of more than 100,000 yuan and less than 500,000 yuan; partial parking), the corresponding consequence score is 4+2=6.

[0132] 2. When there are roads within 20m on both sides of the leak:

[0133] The affected area of ​​the leakage point is 2m 2 The number of people in the highway area is relatively dense, so it is calculated as 1 person per 2 square meters. The highway and its ancillary equipment are calculated at 20,000 yuan / m 2 calculate.

[0134] Personal injury (medium-impact health / safety accident: loss of workdays due to accidental injury; 1 to 2 people slightly injured). Property loss (direct economic loss of less than 100,000 yuan), the corresponding consequence score is 2+1=3.

[0135] 3. When there is a plant water source functional area within 50m on both sides of the leakage point:

[0136] The affected area of ​​the leakage point is 2m 2 The population in key water source functional areas is relatively sparse, and the calculation is 1 person per 10 square meters. The economic value of key water source functional areas is 20,000 yuan / m 2 calculate.

[0137] Personal injury (health / safety accidents with a significant impact: more than 3 people slightly injured, 1 to 2 people seriously injured (including acute industrial poisoning, the same below); exposure exceeds the standard, causing long-term health effects or causing serious occupational diseases). Property loss (direct economic loss of less than 100,000 yuan), the corresponding consequence score is 3+1=4.

[0138] 4. When there are dangerous goods storage places within 50m on both sides of the leakage point:

[0139] The affected area of ​​the leakage point is 2m 2 The dangerous goods storage places are densely populated, and the calculation is 1 person per 2 square meters. The economic value of the dangerous goods factory area is 10,000 yuan / m 2 Due to the special situation of this area, the accident is likely to cause secondary casualties and property losses, and the casualties and property losses will be calculated double.

[0140] Personal injury (major safety accident resulting in death or serious injury: 3 to 9 people seriously injured in the boundary area; 1 to 2 people seriously injured outside the boundary area. Property loss (direct economic loss less than 100,000 yuan), the corresponding consequence score is 4+1=5.

[0141] 5. When there is a high-voltage power transmission area within 50m on both sides of the leakage point:

[0142] The affected area of ​​the leakage point is 2m 2 The population in the high-voltage transmission area is relatively dense, and the calculation is 1 person per 2 square meters. The economic value of the high-voltage transmission area is 10,000 yuan / m 2 Due to the special situation of this area, the accident is likely to cause secondary casualties and property losses, and the casualties and property losses will be calculated double.

[0143] Personal injury (major safety accident resulting in death or serious injury: 3 to 9 people seriously injured in the boundary area; 1 to 2 people seriously injured outside the boundary area). Property loss (direct economic loss less than 100,000 yuan), the corresponding consequence score is 4+1=5.

[0144] The scoring criteria for the consequences of water pipeline failure are shown in the table below.

[0145]

[0146] S63. Calculate the risk of water pipelines in refineries;

[0147] Risk refers to the harm that people bring to humans during a certain period of time when they are engaged in a certain activity. The harm caused to water pipes depends on the probability of an accident and the severity of the consequences of the incident. The risk of water pipes can be defined as the product of the probability of a pipeline leak and the consequences of the leak, specifically R = P × C.

[0148] Where: R is the risk of water pipelines, P is the probability of water pipeline leakage, and C is the consequence index value of water pipeline leakage accidents. S64. Establish a risk index grading table for accidents in water pipelines of refineries.

[0149] The consequences, probability of failure and risk level of water pipeline failure in refining enterprises are shown in the table below. The risk index in the table is the R value calculated above. The value corresponds to a certain level (I, II, III, IV) if it falls within the risk index range.

[0150]

[0151] The above description is only an embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the principles of the present invention should be included in the protection scope of the present invention.

[0152] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0153] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. In addition, the terms "first", "second", etc. are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, features defined as "first", "second", etc. may explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise specified, "multiple" means two or more.

[0154] In the description of the invention, it should be noted that, unless otherwise clearly specified and limited, the terms "install", "connect", "connect", and "set" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the invention can be understood according to specific circumstances.

[0155] Technical features not described in the present invention can be achieved through or by adopting existing technologies and will not be described in detail here. Of course, the above description is not a limitation of the present invention, and the present invention is not limited to the above examples. Changes, modifications, additions or substitutions made by ordinary technicians in this technical field within the essential scope of the present invention should also fall within the scope of protection of the present invention.

Claims

1. A method for risk assessment of water pipelines in refineries based on Bayesian networks, characterized in that: The specific steps are: S1. Collect previous pipeline failure cases and build a historical pipeline failure database; S2. Identify failure factors through historical data in the historical failure database and establish an accident tree based on the failure of the water pipeline of the refinery; S3, determine the prior failure probability of the failure factor and construct a Bayesian network structure; S4, calculating the fuzzy probability of the root node and determining the correction factor; S5. Calculate the a posteriori failure probability of the water pipelines used by the refinery.

2. The method for risk assessment of water pipelines for refineries and petrochemical enterprises based on Bayesian networks according to claim 1 is characterized in that: Step S2 is specifically as follows: S21. Analyze the failure factors of pipeline failure data to determine the basic events and intermediate events that lead to failure; S22. Construct an accident tree according to the relationship between the basic events and the intermediate events.

3. The method for risk assessment of water pipelines for refineries and petrochemical enterprises based on Bayesian networks according to claim 1 is characterized in that: Step S3 is specifically as follows: S31. Determine the prior failure probability of each failure factor based on existing pipeline failure cases; S32, determining the relationship between the parent node and the child node according to the prior failure probability and the established fault tree, and converting the network structure of the fault tree into a Bayesian network structure; S33. Determine the fuzzy subsets of the root node and intermediate node failure probabilities in the Bayesian network.

4. The method for risk assessment of water pipelines for refineries based on Bayesian networks according to claim 1 is characterized in that: Step S4 is specifically as follows: S41. Consider the pipeline, collect the distribution information of the above parameters, and prepare a failure factor classification table, which is divided into primary, secondary and tertiary levels; S42, assigning values ​​to the parameters in the previous step according to the structural importance, and preparing a failure factor weight table; S43, converting each level into a corresponding fuzzy number; S44, converting the fuzzy number into fuzzy probability through a formula; The calculation formula of fuzzy probability is as follows: Among them, FP is fuzzy probability, FPS is fuzzy number, r is average failure probability; the average failure probability of the pipeline system is determined according to the statistical data of the failure database; K and k are the intermediate parameter values ​​of the calculation process; S45. The correction factor of level three (or level two) is set to "1", and the correction factors corresponding to other levels are the ratios of the fuzzy probabilities corresponding to each level to the fuzzy probabilities corresponding to level three (or level two).

5. The method for risk assessment of water pipelines for refineries based on Bayesian networks according to claim 1 is characterized in that: Step S5 is specifically as follows: the failure probability calculation model adopts the following formula: Among them, P i is the pipeline failure probability caused by failure factor i, R i is the prior failure probability corresponding to failure factor i, F i is the failure probability correction factor corresponding to failure factor i.

6. The method for risk assessment of water pipelines for refineries based on Bayesian networks according to claim 1 is characterized in that: The specific steps also include: S6. Calculate the failure risk of water pipelines based on the a posteriori failure probability and establish a classification table for the consequences of pipeline failure.

7. The method for risk assessment of water pipelines for refineries based on Bayesian networks according to claim 6 is characterized in that: The step S6 is specifically as follows: S61. Develop risk assessment standards for various failure factors of water pipelines in refineries; S62. Establish a pipeline failure consequence classification table and failure consequence scoring criteria; S63. Calculate the risk of water pipelines in refineries; S64. Establish a classification table for the likelihood of accidents occurring in water pipelines of refineries and petrochemical enterprises and a classification table for risk index.

8. The method for risk assessment of water pipelines for refineries based on Bayesian networks according to claim 2 is characterized in that: When analyzing the failure factors of pipeline failure data, a comprehensive failure factor analysis is conducted from three aspects: pipeline corrosion, misoperation, and third-party damage. Specifically, it includes pipe age, pipe material, properties of the conveying medium, anti-corrosion layer conditions, pipeline burial depth, pipe diameter, pipeline penetration and span conditions, pipeline laying conditions, and pipeline leakage conditions.

Images