Offshore platform wellhead device dynamic risk assessment method based on evidence change
By establishing a dynamic Bayesian network and matter-element theory model for wellhead devices, and combining it with evidence fusion technology, the problems of integrating new evidence and comprehensively assessing risk probability and failure consequences in existing risk assessment methods have been solved. This has enabled real-time risk assessment and prediction of wellhead devices on offshore platforms, improving safety and reliability.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-12
- Publication Date
- 2026-04-03
AI Technical Summary
Existing risk assessment methods for offshore platform wellhead equipment cannot integrate new evidence in a timely manner and fail to comprehensively consider risk probabilities and failure consequences, resulting in inaccurate and incomplete assessments that cannot effectively support oil and gas production decisions.
A dynamic risk assessment method based on evidence change is adopted. By establishing a dynamic Bayesian network structure model and matter-element theory model of the wellhead device, and combining evidence fusion technology, the risk probability and failure consequences are quantified, enabling real-time analysis and prediction of new evidence.
It enables real-time, dynamic assessment and prediction of wellhead equipment risks, provides strong decision support, improves the safety and reliability of offshore platform wellhead equipment, reduces potential risks, and ensures the smooth operation of oil and gas production.
Smart Images
Figure CN121787892A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of petroleum engineering technology, and more particularly, to a method for dynamic risk assessment of offshore platform wellhead equipment based on changes in evidence. Background Technology
[0002] In the field of oil and gas extraction, the wellhead equipment on offshore platforms is the key equipment at the top of the oil and gas well, bearing the important responsibility of controlling and regulating oil and gas production. Its performance directly determines whether the oil and gas well can achieve safe and efficient production operations. With the continuous development of offshore oil and gas activities, wellhead equipment faces many severe challenges.
[0003] Offshore platform wellhead assemblies operate in harsh marine environments, enduring not only seawater corrosion but also complex operating conditions. Over time, these assemblies develop varying degrees of rust, corrosion, and erosion, weakening their structural strength and sealing performance. Furthermore, areas with localized stress concentrations are more prone to problems during long-term service, with older wellhead assemblies facing a higher risk of failure. Valve leakage and malfunctioning safety valves are also common issues with offshore platform wellhead assemblies, potentially leading to serious safety incidents such as oil and gas leaks, fires, and explosions, resulting in significant safety hazards and economic losses for oil and gas production.
[0004] Given the unique characteristics of the marine environment, offshore platform wellhead equipment has the following features: First, it requires extremely high safety and reliability; even a minor malfunction could lead to catastrophic consequences. Second, it must be highly operable, with easily replaceable parts, enabling rapid repair and maintenance in this special environment. Third, due to its long-term exposure to the marine environment, it requires extremely high corrosion resistance, necessitating special anti-corrosion measures to extend its service life. Furthermore, the wellhead equipment must be compatible with both manual and automatic control to handle different operating scenarios and emergency situations. Its relatively complex structure, integrating multiple functional components, also increases the difficulty and cost of maintenance. Finally, due to its special nature, offshore platform wellhead equipment is expensive, and the cost of repair and replacement in the event of a malfunction is substantial.
[0005] Risk assessment, as a safety evaluation method, is widely used in the oil and gas production sector to evaluate the probability of systemic risk events and the potential consequences. Through risk assessment, potential safety hazards can be identified in advance, allowing for corresponding measures to mitigate risks and improve system safety. However, current risk assessment methods have certain limitations. These methods often fail to consider the impact of emerging evidence on the risk assessment results. In actual oil and gas production, new data and information constantly emerge over time and with equipment operation. This new evidence may alter the original risk assessment results, but existing methods cannot effectively integrate and analyze this new evidence in a timely manner. Furthermore, existing risk assessment methods do not fully reflect the combined impact of risk probability and failure consequences. The assessment process typically focuses only on the probability of risk occurrence or the severity of failure consequences, without combining both for a comprehensive evaluation. This may lead to inaccurate and incomplete risk assessments, failing to provide strong support for oil and gas production decisions.
[0006] Therefore, considering the unique characteristics of offshore platform wellhead equipment and the problems with existing risk assessment methods, developing a dynamic risk assessment method that can take into account the impact of new evidence and comprehensively assess the probability of risk and the consequences of failure is of great practical significance for improving the safety and reliability of offshore platform wellhead equipment and ensuring the smooth operation of oil and gas production. Summary of the Invention
[0007] This invention aims to address at least one of the technical problems existing in the prior art. To this end, this invention provides a dynamic risk assessment method for offshore platform wellhead devices based on evidence changes, aiming to achieve risk assessment and prediction of wellhead devices, thereby providing decision support for risk management by relevant units and ensuring the safe and reliable operation of the wellhead device system.
[0008] To achieve the above objectives, the present invention adopts the following technical solution: a dynamic risk assessment method for offshore platform wellhead devices based on evidence changes, comprising the following steps: Based on data on the structural composition and failure rate of wellhead equipment, a dynamic Bayesian network structure model of wellhead equipment is established to determine failure modes, key components, subsystems, and logical relationships between systems. Based on the established dynamic Bayesian network parameter model of the wellhead device, the relationship between network nodes is quantified through state discretization, conditional probability calculation, etc. Establish a failure consequence assessment model based on matter-element theory to determine the failure mode, key components, subsystems, and quantitative values of system failure consequences; The newly added evidence to the wellhead device is transformed into a risk range. After conflict value quantification and evidence fusion, the risk range of the wellhead device after the appearance of the newly added evidence is calculated, and the risk is predicted.
[0009] As a preferred embodiment, the method for establishing the dynamic Bayesian network structure model of the wellhead device is as follows: Identify the key components of the wellhead assembly; Identify the main failure modes of each key component; Establish a priori network model that considers subsystems and key components; Based on the time relationship of each node in the fault mode layer of the prior network model, it is extended into a transition network model, and the prior network model and the transition network model together form a dynamic Bayesian network structure model of the wellhead device.
[0010] As a preferred embodiment, the key components of the wellhead assembly include the main production valve, the production wing valve, the casing wing valve, the wax removal valve, the wellhead cap, the wellhead body, the throttle valve, and the surface safety valve. The main failure modes of production main valves, production wing valves, sleeve wing valves, wax removal valves, throttle valves, and ground safety valves are classified into switch failures, external valve leakage, internal valve leakage, and other failures. The main causes of switch failure include: impurities and dirt buildup at the valve stem thread connection; corrosive components in the medium and the reaction of CO2, O2 and moisture; bearing corrosion; The main causes of valve leakage include: failure of valve stem seals due to aging, wear, cracking, gap seizure, or permanent deformation of the sealing rings; failure of the steel ring seal at the connection between the valve body and the valve cover due to insufficient clamping force caused by corrosion of the steel ring sealing surface, loosening or unfastening of flange connection bolts; insufficient clamping force caused by reduced elasticity of the spring inside the grease injection valve; corrosion of the steel ball; and gaps formed by impurities and dirt stuck on the sealing surface, leading to grease injection valve sealing failure. Internal leakage in valves is generally caused by the failure of the seal between the valve seat and the valve plate. The main causes include: impurity accumulation, acid corrosion, wear of the sealing ring, and loss of sealing grease. The main failures of the oil well cap are structural defects, external leakage, and other malfunctions. The main failures of the wellhead are body cracking, thinning, and other defects; Other malfunctions refer to those caused by human error, such as improper operation or incomplete management systems.
[0011] Preferably, the prior network model considering the subsystem is divided into four layers from top to bottom: the first layer is the failure mode layer, where the nodes contain all failure modes of a single component; the second layer is the critical component layer, where the nodes represent the state of each component of the subsystem under the influence of different failure modes; the third layer is the subsystem layer, where the nodes represent the state of the wellhead device subsystem under the influence of multiple identical components; and the fourth layer is the system layer, where the nodes represent the state of the wellhead device under the combined influence of the subsystem and critical components. The prior network model considering key components has three layers: the first layer is the failure mode layer, the second layer is the key component layer, and the third layer is the system layer. Compared with the Bayesian network model considering subsystems, it lacks the subsystem layer, but the rest of the configuration is the same.
[0012] As a preferred embodiment, the method for establishing the dynamic Bayesian network parameter model of the wellhead device is as follows: All nodes in the established dynamic Bayesian network structure model of the wellhead device are discretized, that is, all nodes are divided into five states, from best to worst: "perfect", "good", "average", "poor" and "very poor". Based on reliability data and wellhead equipment failure statistics, the lower limit of the failure rate for each failure mode layer node is given. l min and failure rate limit l max ; Based on the discretization results of each node's state and the upper and lower limits of the failure rate, calculate the probability of each node in the failure mode layer failing over time. By improving the Leaky Noisy-or model and using Gaussian distribution to jointly calculate the conditional probability of the prior network, a dynamic Bayesian network parameter model for the wellhead device is established.
[0013] Preferably, the formula for calculating the probability of failure is:
[0014] In the formula, l ij For fault mode layer nodes by i The state becomes j The failure rate in the state, and l max >l ij >l min , i, j =1, 2, 3, 4, 5, representing "perfect", "good", "average", "poor" and "very poor" status respectively; The conditional probability of the prior network is derived by generalizing the following formula:
[0015] In the formula, X t0 i and X t0 j They are respectively t 0-time node X i and X j State combinations, X i and X j The first i and the j Each node.
[0016] As a preferred embodiment, the method for establishing the wellhead device failure consequence assessment model based on matter-element theory is as follows: For the failure consequence assessment of wellhead equipment, five aspects are classified into five levels: functional impact, economic loss, personnel impact, environmental impact, and other impacts, in ascending order of severity. N 1. N 2. N 3. N 4 and N 5; Therefore, a failure mode consequence level set can be established. N ={ N 1. N 2. N 3. N 4. N 5} and failure consequence feature set c ={ c 1, c 2, c 3, c 4, c 5} = {Functional impact, economic loss, personnel impact, environmental impact, other impacts}; Based on the established failure mode consequence level set and failure consequence feature set, the classical domain matrix for wellhead equipment failure consequence assessment is calculated. R 1-5 and section R u ; Based on expert experience and wellhead equipment failure cases, the following material elements are identified as having the consequences of each failure mode of the wellhead equipment:
[0017] In the formula, c i For the failure consequence feature set; Vi For failure consequence feature set c i The value; The comparison matrix constructed using the analytic hierarchy process (AHP) is used to determine the characteristic sets of different failure consequences. Determine the association function K j ( V i ), comprehensive correlation K j ( P Failure Consequence Level F j and quantification of failure consequences F， Complete the failure consequence assessment based on matter-element theory:
[0018]
[0019]
[0020]
[0021] In the formula, V ji Feature set of failure consequences in classical domain c i The range of values; P ( V i , V ji )for V i and V ji The distance between intervals.
[0022] As a preferred option: the classical domain matrix for assessing the failure consequences of the wellhead device system. R 1-5 and section R u The calculation formulas are as follows:
[0023]
[0024] In the formula, u For all consequence levels in the failure mode consequence level set; V u For the failure consequence feature set in the segment domain c i The range of values; The comparison matrix constructed by the analytic hierarchy process is as follows:
[0025] From the comparison matrix B It can be seen that: Consistency index CI =0.091, random consistency index RI =1.11, Consistency Ratio CR = CI / RI =0.0820 < 0.1, which meets the consistency test, and the weight matrix is obtained as follows:
[0026] In the formula, p i The weights are the feature sets of failure consequences.
[0027] As a preferred embodiment, the method for quantifying the newly added evidence conflict value and fusing evidence in the wellhead device is as follows: Based on risk information of wellhead equipment obtained from different physical levels, different characteristics and different lifespans, including the status rating of each key component, the mean time to failure of each key component, and the failure rate of each key component's failure mode; The obtained risk information is combined with system state discretization and knowledge of risk domains, and all of it is transformed into risk intervals. R i =[ R min R max This, together with the risk range inferred through the dynamic Bayesian network structure model of the wellhead device, constitutes multiple pieces of evidence; To facilitate the quantification and fusion of conflict values of evidence, the risk range is expressed as a percentage, using the following formula:
[0028] In the formula, R min and R max These are the minimum risk value and the maximum risk value, respectively. The risk percentage value range is transformed into the trust function and likelihood function of the evidence theory using the following formula:
[0029] In the formula, Bell ( X ) represents "Jiao Yuan" X The lowest confidence interval for "true"; Pl ( X ) represents "Jiao Yuan" X The highest confidence interval for "true"; The quality function of risk in wellhead equipment operation is calculated using the trust function and the likelihood function. By calculating the joint conflict factor and fusing the evidence, a new risk range is obtained. Then, the state of each failure mode is solved by the least squares method to obtain the prior probability of DBN, which can be used to predict and diagnose the risk.
[0030] Preferably, the formula for calculating the mass function is as follows:
[0031] In the formula, This indicates a certain possibility of risk. This indicates a possibility that there is absolutely no risk. Indicates the range of uncertainty related to risk; It is an empty set.
[0032] The present invention has the following advantages due to the adoption of the above technical solutions: 1. This invention ingeniously integrates Deep Belief Networks (DBN) and matter-element theory to accurately quantify the risk probability and failure consequences of offshore wellhead equipment. By introducing evidence theory and joint conflict factors, it effectively solves the problem of fusing newly emerging evidence and constructs a dynamic risk assessment method model for offshore wellhead equipment based on evidence changes. This innovative method can not only conduct real-time and dynamic risk assessment of wellhead equipment systems but also achieve risk prediction, providing strong decision support for risk management by relevant units. This ensures the safe and reliable operation of wellhead equipment systems and greatly improves the safety and reliability of offshore oil and gas extraction.
[0033] 2. This invention provides an in-depth analysis of the failure modes of key components of the wellhead system over time, accurately calculates their risk probabilities and failure consequences, and derives corresponding risk ranges. This process not only encompasses a comprehensive risk analysis of key components but also resolves the conflict judgment problem between newly emerging evidence through an innovative fusion method, achieving effective evidence integration. This enables the invention to more accurately assess the overall risk status of the wellhead system, providing a scientific basis for the maintenance and management of the wellhead system during offshore oil and gas extraction, effectively reducing potential risks caused by key component failures, and ensuring the smooth progress of offshore oil and gas extraction.
[0034] In summary, this invention has significant innovative value and practical significance in the field of risk assessment for offshore wellhead equipment, and provides advanced technical support for the safety management of offshore oil and gas extraction. Attached Figure Description
[0035] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts. In the drawings: Figure 1 A flowchart of the dynamic risk assessment method for offshore platform wellhead devices provided by the present invention; Figure 2 This is a dynamic Bayesian network structure model for wellhead equipment. Detailed Implementation
[0036] To make the objectives, technical solutions, and advantages of the present invention clearer, specific embodiments of the present invention will be further described below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.
[0037] This invention provides a dynamic risk assessment method for offshore platform wellhead equipment based on evidence changes, comprising the following steps: establishing a dynamic Bayesian network structure model of the wellhead equipment based on its structural composition and failure rate; quantifying the relationships between network nodes through state discretization and conditional probability calculation based on the established dynamic Bayesian network parameter model; establishing a failure consequence assessment model based on matter-element theory to determine the quantitative values of failure modes, key components, subsystems, and system failure consequences; converting newly added evidence for the wellhead equipment into risk intervals, and calculating the risk interval of the wellhead equipment after the appearance of newly added evidence through conflict value quantification and evidence fusion, and predicting the risk. This invention can realize risk assessment and risk prediction for wellhead equipment, aiming to provide decision support for the risk management work of relevant units and ensure the safe and reliable operation of the wellhead equipment system.
[0038] The following is a detailed description, with reference to the accompanying drawings, of the dynamic risk assessment method for offshore platform wellhead devices based on evidence changes provided by the embodiments of the present invention.
[0039] Please see Figure 1 This embodiment provides a dynamic risk assessment method for offshore platform wellhead devices based on changes in evidence, including the following steps: S100. Based on data such as the structural composition and failure rate of the wellhead equipment, a dynamic Bayesian network structure model of the wellhead equipment is established to determine the failure modes, key components, subsystems, and logical relationships between the systems. The method for establishing the dynamic Bayesian network structure model of the wellhead equipment is as follows: S101. Identify the key components of the wellhead equipment. The wellhead equipment mainly refers to the part above the wellhead main gate valve. Key components include the main production valve, production wing valve, casing wing valve, wax removal valve, wellhead cap, wellhead body, throttle valve, surface safety valve, etc. S102. Determine the main failure modes of each key component. For valves such as production main valve, production wing valve, sleeve wing valve, wax removal valve, throttle valve and ground safety valve, the main failure modes can be divided into switch failure, valve external leakage, valve internal leakage and other failures. The main causes of switch failure include: impurities and dirt accumulation at the valve stem thread connection; the reaction of corrosive components in the medium (such as H2S, various acids, etc.) and CO2, O2, etc. with moisture; and bearing corrosion. The main causes of valve leakage include: failure of valve stem sealing rings due to aging, wear, cracking, gap seizure, permanent deformation, etc.; failure of steel ring seal at the connection between valve body and valve cover due to corrosion of steel ring sealing surface, loosening or unfastening of flange connection bolts leading to insufficient clamping force; failure of grease injection valve seal due to reduced elasticity of spring in grease injection valve causing insufficient clamping force, corrosion of steel ball, and impurities and dirt stuck on the sealing surface forming gaps. Internal leakage in valves is generally caused by the failure of the seal between the valve seat and the valve plate. The main causes include: impurity accumulation, acid corrosion, wear of the sealing ring, and loss of sealing grease. Other malfunctions refer to those caused by human error, such as improper operation or incomplete management systems. The main failures of the oil well cap are structural defects, external leakage, and other malfunctions. The main failures of the wellhead are body cracking, thinning, and other defects; S103. Establish a priori network model that considers subsystems and key components; The prior network model considering the subsystem is divided into four layers from top to bottom. The first layer is the failure mode layer, where the nodes contain all failure modes of a single component. The second layer is the critical component layer, where the nodes represent the state of each component of the subsystem under the influence of different failure modes. The third layer is the subsystem layer, where the nodes represent the state of the wellhead device subsystem under the influence of multiple identical components. The fourth layer is the system layer, where the nodes represent the state of the wellhead device under the combined influence of the subsystem and critical components. The prior network model considering key components has three layers: the first layer is the failure mode layer, the second layer is the key component layer, and the third layer is the system layer. Compared with the Bayesian network model considering subsystems, it lacks the subsystem layer, but the rest of the configuration is the same. S104. Based on the temporal relationships of each node in the fault mode layer of the prior network model, it is extended into a transition network model. The prior network model and the transition network model together form a dynamic Bayesian network structure model for the wellhead device (see [link]). Figure 2 ).
[0040] S200. Based on the established dynamic Bayesian network parameter model of the wellhead device, the relationships between network nodes are quantified through state discretization, conditional probability calculation, etc.; the method for establishing the dynamic Bayesian network parameter model of the wellhead device is as follows: S201. Discretize all nodes in the established dynamic Bayesian network structure model of the wellhead device (including all nodes in the fault mode layer, key component layer, subsystem layer, and system layer), that is, all nodes are divided into five states, from best to worst: "perfect", "good", "average", "poor" and "extremely poor". S202. Based on reliability data and wellhead equipment failure statistics, provide the lower limit of the failure rate for each failure mode layer node. l min and failure rate limit l max ; S203. Based on the discretization results of each node's state and the upper and lower limits of the failure rate, calculate the probability of each node in the failure mode layer failing over time; whereby the probability calculation formula is:
[0041] In the formula, l ij For fault mode layer nodes by i The state becomes j The failure rate in the state, and l max >l ij >l min , i, j =1, 2, 3, 4, 5, representing "perfect", "good", "average", "poor" and "very poor" states, respectively.
[0042] S204. By improving the Leaky Noisy-or model and using the Gaussian distribution to jointly calculate the conditional probabilities of the prior network, a dynamic Bayesian network parameter model for the wellhead device is established. The conditional probabilities of the prior network are derived by generalizing the following formula:
[0043] In the formula, X t0 i and X t0 j They are respectively t 0-time node X i and Xj State combinations, X i and X j The first i and the j Each node.
[0044] S300. Establish a failure consequence assessment model based on matter-element theory to determine the failure modes, key components, subsystems, and quantitative values of system failure consequences; the method for establishing the wellhead device failure consequence assessment model based on matter-element theory is as follows: S301. For the failure consequence assessment of wellhead equipment, five aspects are classified into five levels: functional impact, economic loss, personnel impact, environmental impact, and other impacts, in ascending order of severity: N 1. N 2. N 3. N 4 and N 5. Therefore, a failure mode consequence level set can be established. N ={ N 1. N 2. N 3. N 4. N 5} and failure consequence feature set c ={ c 1, c 2, c 3, c 4, c 5} = {Functional impact, economic loss, personnel impact, environmental impact, other impacts}; S302. Based on the established failure mode consequence level set and failure consequence feature set, the classical domain matrix for wellhead equipment failure consequence assessment is calculated. R 1-5 and section R u Among them, the classical domain matrix for assessing the failure consequences of wellhead equipment systems. R 1-5 and section R u The calculation formulas are as follows:
[0045]
[0046] In the formula, u For all consequence levels in the failure mode consequence level set; V u For the failure consequence feature set in the segment domain c i The range of values.
[0047] S303. Based on expert experience and wellhead equipment failure cases, determine the evaluation material elements for each failure mode of the wellhead equipment;
[0048] In the formula, c i For the failure consequence feature set; V i For failure consequence feature set c i The value of .
[0049] S304. The comparison matrix constructed using the analytic hierarchy process (AHP) is used to determine the characteristic sets of different failure consequences; the comparison matrix constructed using the AHP is as follows:
[0050] From the comparison matrix B It can be seen that: Consistency index CI =0.091, random consistency index RI =1.11, Consistency Ratio CR = CI / RI =0.0820 < 0.1, which meets the consistency test, and the weight matrix is obtained as follows:
[0051] In the formula, p i The weights are the feature sets of failure consequences.
[0052] S305. Determine the correlation function K j ( V i ), comprehensive correlation K j ( P Failure Consequence Level F j and quantification of failure consequences F， Complete the failure consequence assessment based on matter-element theory:
[0053]
[0054]
[0055]
[0056] In the formula, V jiFeature set of failure consequences in classical domain c i The range of values; P ( V i , V ji )for V i and V ji The distance between intervals.
[0057] S400. The newly added evidence to the wellhead device is converted into a risk range. After conflict value quantification and evidence fusion, the risk range of the wellhead device after the appearance of the newly added evidence is calculated, and the risk is predicted. The methods for conflict value quantification and evidence fusion of the newly added evidence to the wellhead device are as follows: S401. Risk information of wellhead equipment is obtained from multiple aspects such as different physical levels, different characteristics and different lifespans, including the status rating of each key component, the average failure time of each key component, and the failure rate of each key component's failure mode. S402. Combine the risk information obtained in step S401 with the knowledge of system state discretization and risk domain, and transform it into risk intervals. R i =[ R min R max This, together with the risk range inferred through the dynamic Bayesian network structure model of the wellhead device, constitutes multiple pieces of evidence; S403. To facilitate the quantification and fusion of conflict values of evidence, the risk interval is expressed as a percentage, and the formula is as follows:
[0058] In the formula, R min and R max These are the minimum risk value and the maximum risk value, respectively. S404. The risk percentage value range is transformed into the trust function and likelihood function of the evidence theory using the following formula:
[0059] In the formula, Bell ( X ) represents "Jiao Yuan" X The lowest confidence interval for "true"; Pl ( X ) represents "Jiao Yuan" X The highest confidence interval for "true"; S405. The quality function of the risk during wellhead equipment operation is calculated using the trust function and the likelihood function; the formula for calculating the quality function is as follows:
[0060] In the formula, This indicates a certain possibility of risk. This indicates a possibility that there is absolutely no risk. Indicates the range of uncertainty related to risk; It is an empty set; S406. Calculate the joint conflict factor and fuse the evidence to obtain a new risk range. Then, use the least squares method to solve for the state of each failure mode to obtain the prior probability of DBN in 12 years. This allows for risk prediction and diagnosis.
[0061] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and not to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. These modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and they should all be covered within the scope of the claims and specification of this application. In particular, as long as there is no structural conflict, the various technical features mentioned in the embodiments can be combined in any way. This application is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.
Claims
1. A dynamic risk assessment method for offshore platform wellhead equipment based on changes in evidence, characterized in that, Includes the following steps: Based on data on the structural composition and failure rate of wellhead equipment, a dynamic Bayesian network structure model of wellhead equipment is established to determine failure modes, key components, subsystems, and logical relationships between systems. Based on the established dynamic Bayesian network parameter model of the wellhead device, the relationship between network nodes is quantified through state discretization, conditional probability calculation, etc. Establish a failure consequence assessment model based on matter-element theory to determine the failure mode, key components, subsystems, and quantitative values of system failure consequences; The newly added evidence to the wellhead device is transformed into a risk range. After conflict value quantification and evidence fusion, the risk range of the wellhead device after the appearance of the newly added evidence is calculated, and the risk is predicted.
2. The dynamic risk assessment method for offshore platform wellhead equipment according to claim 1, characterized in that, The method for establishing the dynamic Bayesian network structure model of the wellhead device is as follows: Identify the key components of the wellhead assembly; Identify the main failure modes of each key component; Establish a priori network model that considers subsystems and key components; Based on the time relationship of each node in the fault mode layer of the prior network model, it is extended into a transition network model, and the prior network model and the transition network model together form a dynamic Bayesian network structure model of the wellhead device.
3. The dynamic risk assessment method for offshore platform wellhead equipment according to claim 2, characterized in that, The key components of the wellhead equipment include the main production valve, production wing valve, casing wing valve, wax removal valve, wellhead cap, wellhead body, throttle valve, and surface safety valve. The main failure modes of production main valves, production wing valves, sleeve wing valves, wax removal valves, throttle valves, and ground safety valves are classified into switch failures, external valve leakage, internal valve leakage, and other failures. The main causes of switch failure include: impurities and dirt buildup at the valve stem thread connection; corrosive components in the medium and the reaction of CO2, O2 and moisture; bearing corrosion; The main causes of valve leakage include: failure of valve stem seals due to aging, wear, cracking, gap seizure, or permanent deformation of the sealing rings; failure of the steel ring seal at the connection between the valve body and the valve cover due to insufficient clamping force caused by corrosion of the steel ring sealing surface, loosening or unfastening of flange connection bolts; insufficient clamping force caused by reduced elasticity of the spring inside the grease injection valve; corrosion of the steel ball; and gaps formed by impurities and dirt stuck on the sealing surface, leading to grease injection valve sealing failure. Internal leakage in valves is generally caused by the failure of the seal between the valve seat and the valve plate. The main causes include: impurity accumulation, acid corrosion, wear of the sealing ring, and loss of sealing grease. The main failures of the oil well cap are structural defects, external leakage, and other malfunctions. The main failures of the wellhead are body cracking, thinning, and other defects; Other malfunctions refer to those caused by human error, such as improper operation or incomplete management systems.
4. The dynamic risk assessment method for offshore platform wellhead equipment according to claim 3, characterized in that, The prior network model considering the subsystem is divided into four layers from top to bottom. The first layer is the failure mode layer, where the nodes contain all failure modes of a single component. The second layer is the critical component layer, where the nodes represent the state of each component of the subsystem under the influence of different failure modes. The third layer is the subsystem layer, where the nodes represent the state of the wellhead device subsystem under the influence of multiple identical components. The fourth layer is the system layer, where the nodes represent the state of the wellhead device under the combined influence of the subsystem and critical components. The prior network model considering key components has three layers: the first layer is the failure mode layer, the second layer is the key component layer, and the third layer is the system layer. Compared with the Bayesian network model considering subsystems, it lacks the subsystem layer, but the rest of the configuration is the same.
5. The dynamic risk assessment method for offshore platform wellhead equipment according to claim 4, characterized in that, The method for establishing the dynamic Bayesian network parameter model of the wellhead device is as follows: All nodes in the established dynamic Bayesian network structure model of the wellhead device are discretized, that is, all nodes are divided into five states, from best to worst: "perfect", "good", "average", "poor" and "very poor". Based on reliability data and wellhead equipment failure statistics, the lower limit of the failure rate for each failure mode layer node is given. λ min and failure rate limit λ max ; Based on the discretization results of each node's state and the upper and lower limits of the failure rate, calculate the probability of each node in the failure mode layer failing over time. By improving the Leaky Noisy-or model and using Gaussian distribution to jointly calculate the conditional probability of the prior network, a dynamic Bayesian network parameter model for the wellhead device is established.
6. The dynamic risk assessment method for offshore platform wellhead equipment according to claim 5, characterized in that, The formula for calculating the probability of failure is: In the formula, λ ij For fault mode layer nodes by i The state becomes j The failure rate in the state, and λ max >λ ij >λ min , i, j =1, 2, 3, 4, 5, representing "perfect", "good", "average", "poor" and "very poor" states respectively; The conditional probability of the prior network is derived by generalizing the following formula: In the formula, X t0 i and X t0 j They are respectively t 0-time node X i and X j State combinations, X i and X j The first i and the j Each node.
7. The dynamic risk assessment method for offshore platform wellhead equipment according to claim 6, characterized in that, The method for establishing the wellhead device failure consequence assessment model based on matter-element theory is as follows: For the failure consequence assessment of wellhead equipment, five aspects are classified into five levels: functional impact, economic loss, personnel impact, environmental impact, and other impacts, in ascending order of severity. N 1. N 2. N 3. N 4 and N 5; Therefore, a failure mode consequence level set can be established. N ={ N 1. N 2. N 3. N 4. N 5} and failure consequence feature set c ={ c 1, c 2, c 3, c 4, c 5} = {Functional impact, economic loss, personnel impact, environmental impact, other impacts}; Based on the established failure mode consequence level set and failure consequence feature set, the classical domain matrix for wellhead equipment failure consequence assessment is calculated. R 1-5 and section R u ; Based on expert experience and wellhead equipment failure cases, the following material elements are identified as having the consequences of each failure mode of the wellhead equipment: In the formula, c i For the failure consequence feature set; V i For failure consequence feature set c i The value; The comparison matrix constructed using the analytic hierarchy process (AHP) is used to determine the characteristic sets of different failure consequences. Determine the association function K j ( V i ), comprehensive correlation K j ( P Failure Consequence Level F j and quantification of failure consequences F， Complete the failure consequence assessment based on matter-element theory: In the formula, V ji Feature set of failure consequences in classical domain c i The range of values; P ( V i , V ji )for V i and V ji The distance between intervals.
8. The dynamic risk assessment method for offshore platform wellhead equipment according to claim 7, characterized in that, Classical domain matrix for assessing the failure consequences of the wellhead device system R 1-5 and section R u The calculation formulas are as follows: In the formula, u For all consequence levels in the failure mode consequence level set; V u For the failure consequence feature set in the segment domain c i The range of values; The comparison matrix constructed by the analytic hierarchy process is as follows: From the comparison matrix B It can be seen that: Consistency index CI =0.091, random consistency index RI =1.11, Consistency Ratio CR = CI / RI =0.0820 < 0.1, which meets the consistency test, and the weight matrix is obtained as follows: In the formula, p i The weights are the feature sets of failure consequences.
9. The dynamic risk assessment method for offshore platform wellhead equipment according to claim 8, characterized in that, The methods for quantifying evidence conflict values and fusing evidence newly added to the wellhead device are as follows: Based on risk information of wellhead equipment obtained from different physical levels, different characteristics and different lifespans, including the status rating of each key component, the mean time to failure of each key component, and the failure rate of each key component's failure mode; The obtained risk information is combined with system state discretization and knowledge of risk domains, and all of it is transformed into risk intervals. R i =[ R min R max This, together with the risk range inferred through the dynamic Bayesian network structure model of the wellhead device, constitutes multiple pieces of evidence; To facilitate the quantification and fusion of conflict values of evidence, the risk range is expressed as a percentage, using the following formula: In the formula, R min and R max These are the minimum risk value and the maximum risk value, respectively. The risk percentage value range is transformed into the trust function and likelihood function of the evidence theory using the following formula: In the formula, Bel ( X ) represents "Jiao Yuan" X The lowest confidence interval for "true"; Pl ( X ) represents "Jiao Yuan" X The highest confidence interval for "true"; The quality function of risk in wellhead equipment operation is calculated using the trust function and the likelihood function. By calculating the joint conflict factor and fusing the evidence, a new risk range is obtained. Then, the state of each failure mode is solved by the least squares method to obtain the prior probability of DBN, which can be used to predict and diagnose the risk.
10. The dynamic risk assessment method for offshore platform wellhead equipment according to claim 9, characterized in that, The formula for calculating the mass function is as follows: In the formula, This indicates a certain possibility of risk. This indicates a possibility that there is absolutely no risk. Indicates the range of uncertainty related to risk; It is an empty set.