Multi-model consistency reliability evaluation method, device and medium for underwater vertical christmas tree system

Abstract

The application discloses a multi-model consistency reliability evaluation method, equipment and medium for a subsea vertical Christmas tree system, and belongs to the technical field of reliability evaluation of marine engineering equipment. The method is characterized in that the subsea vertical Christmas tree system is divided into structures, a fuzzy fault tree model, a state space model and a probability network model are respectively established, the reliability level of the system under different modeling assumptions is analyzed and evaluated, and the reliability results of the outputs of the models are obtained under the same working condition. An information entropy weighted fusion method based on information uncertainty is introduced to comprehensively process the multi-model evaluation results, so that stable and comparable system comprehensive reliability indexes are obtained. The application can effectively reduce the influence of a single model assumption on the reliability evaluation result, improve the stability and engineering applicability of the reliability evaluation result of the subsea vertical Christmas tree system, is suitable for structure scheme comparison and selection and reliability analysis under complex working conditions, and has good engineering application value.

Description

Technical Field

[0001] This invention belongs to the field of quantitative assessment methods for the structural reliability of marine engineering equipment, specifically involving a multi-model consistency reliability assessment method, equipment, and medium for underwater vertical production tree systems. This method integrates the modeling advantages of fuzzy fault trees, Markov models, and Bayesian networks, and proposes an adaptive weighting algorithm based on information entropy to dynamically fuse and optimize the reliability indicators output by multiple models. Background Technology

[0002] As offshore oil and gas exploration continues to extend in depth, subsea production systems inevitably face challenges such as complex structures, complex operating conditions (e.g., water depth, working pressure, and marine environmental conditions), and difficulties in installation and maintenance. Therefore, higher demands are placed on the reliability of these equipment. As the core equipment for controlling well production, the subsea wellhead directly impacts the overall oil and gas production status of the subsea system; thus, its structural reliability, safety, and long-term availability are increasingly becoming a focus of attention.

[0003] Because subsea production trees operate in complex marine environments for extended periods, they not only face high pressure, low temperature, and strong corrosion, but are also affected by factors such as operating depth and oil and gas erosion, making them prone to various failure modes such as leakage, corrosion, blockage, and seal failure.

[0004] Existing research focuses on analytical methods based on fault diagnosis or risk prediction, aiming to identify the probability of fault occurrence or risk level, rather than for reliability consistency assessment of different structural schemes and configuration schemes in the design stage. This makes it difficult to meet the quantitative decision-making needs of the subsea production tree engineering design and scheme comparison stage.

[0005] Therefore, this invention proposes a reliability assessment method for subsea production tree systems that combines multiple modeling techniques and is adapted to the complex environment of the deep sea. This method is used to conduct consistency analysis and comprehensive evaluation of the reliability results of subsea production tree systems under different operating condition assumptions and model assumptions during the design and operation and maintenance assessment phases. Summary of the Invention

[0006] The purpose of this invention is to provide a multi-model consistency reliability assessment method, equipment, and medium for subsea vertical tree systems. By introducing multiple reliability modeling methods and constructing a result fusion mechanism based on information uncertainty, the reliability indicators output by different models are comprehensively evaluated, thereby reducing the impact of model assumption differences on engineering decisions and improving the stability and engineering applicability of the reliability assessment results of subsea tree systems.

[0007] The multi-model consistency reliability assessment method for subsea vertical production tree systems described in this invention includes system structure partitioning, multi-model reliability modeling, and model result fusion assessment steps. Specifically, the subsea vertical production tree system is structurally decomposed into a throttling module, a production tree body module, a hydraulic connector module, and a tubing suspension module. Based on reliability statistics and engineering experience parameters, a fuzzy fault tree model to describe the functional failure relationships of the system, a state space model to describe the state evolution characteristics of the system or subsystems, and a probabilistic network model to describe the structural dependencies of system components are established. Under the same assessment conditions, the reliability assessment results output by the fuzzy fault tree model, the state space model, and the probabilistic network model are obtained, and the assessment results are normalized. The corresponding information entropy value is calculated based on the uncertainty characteristics of each model's output results, and model fusion weights are constructed based on the information entropy. The multi-model reliability assessment results are then weighted and integrated to obtain a comprehensive reliability assessment index for the subsea vertical production tree system.

[0008] Furthermore, the system structure is divided into system level, equipment level and component level. The system level is the subsea vertical Christmas tree system, the equipment level is the throttling module, the Christmas tree body module, the hydraulic connector module and the tubing suspension module, and the component level includes valves, sealing components, hydraulic joints and connecting components.

[0009] Furthermore, the fuzzy fault tree model uses fuzzy set theory to describe the failure probability of basic events in an interval manner, and obtains the system reliability interval and the importance ranking results of key components through fuzzy logic operations.

[0010] Furthermore, the state-space model is a Markov state model, and the steady-state availability of the system or subsystem under long-term operating conditions is calculated through state transition probabilities or state transition rate parameters.

[0011] Furthermore, in the probabilistic network model, nodes are used to represent the operating state of system components or subsystems, the connection relationships between nodes are used to describe the dependencies at the structural or functional level, and the probabilistic network model is used to complete the reliability probability propagation calculation.

[0012] Furthermore, the information entropy value is calculated based on the dispersion or uncertainty characteristics of the reliability assessment results output by each model. The smaller the information entropy value, the greater the weight obtained by the model in the fusion process.

[0013] Furthermore, the comprehensive reliability assessment index is used to compare and analyze the reliability results of subsea vertical production tree systems under different structural configuration schemes or different operating condition assumptions.

[0014] Furthermore, the multi-model consistency reliability assessment process does not involve system fault diagnosis, state identification, or real-time risk prediction processes.

[0015] The electronic device of the present invention includes a processor and a memory, wherein the memory stores computer-executable instructions, and the processor, when executing the instructions, implements the multi-model consistency reliability assessment method for subsea vertical production tree systems.

[0016] The computer-readable storage medium of the present invention stores a computer program, which, when executed by a processor, is used to implement the multi-model consistency reliability assessment method for subsea vertical production tree systems.

[0017] Compared with existing technologies, the present invention exhibits the following beneficial effects in terms of technical performance and practical application: 1. This invention introduces multiple reliability modeling methods to comprehensively analyze the reliability results of subsea vertical production tree systems under different modeling assumptions, effectively reducing the impact of a single model assumption on the evaluation results and improving the stability of engineering evaluation results.

[0018] 2. This invention focuses on the reliability analysis needs in the engineering design and operation and maintenance evaluation stages. It can be used for reliability comparison analysis under different structural configuration schemes or operating condition assumptions, providing a quantitative reference for the engineering design and evaluation of subsea production tree systems.

[0019] 3. By constructing a fusion mechanism based on information uncertainty, the consistency processing of reliability results from multiple models can be achieved, enabling the comprehensive evaluation index to reflect the effective information content of different models and improving the engineering applicability of reliability evaluation results.

[0020] 4. This method does not rely on real-time monitoring or fault identification processes, and is suitable for deep-water engineering scenarios with limited data or incomplete statistical information. It has good versatility and engineering promotion value. Attached Figure Description

[0021] Figure 1 Fault tree diagram of subsea production tree system; Figure 2 Markov model for subsea production tree system; Figure 3 This is a probability network diagram of oil well trees. Detailed Implementation

[0022] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0023] Example 1: The multi-model consensus reliability assessment method for subsea vertical wellhead systems described in this invention includes the following steps: Step 1: System Structure Decomposition and Data Preparation The subsea vertical production tree was structurally and functionally decomposed. By reviewing documents and structural drawings, the functional units and key components of the production tree were identified, dividing the system into four subsystems: a throttling module, a production tree body module, a hydraulic connector module, and a tubing suspension module. Based on relevant reliability data from the DNVOREDA database for subsea systems, failure statistics for the subsea production tree system subsystems were collected. The main failure modes causing production tree failures were identified as leakage, unlocking / locking failure, opening / closing failure, blockage, and corrosion. The primary cause of production tree failure was control system failure, with a significantly higher failure rate than other subsystems. This was followed by piping system failure, tubing suspension failure, throttling module failure, hydraulic connector failure, tree structure and monitoring system failure, and other system failures.

[0024] Based on the engineering structural characteristics of the system, this invention divides the subsea production tree into four subsystems: a throttling module, a production tree body module, a hydraulic connector module, and a tubing suspension module. These four subsystems are considered as the system level, the components contained in the four subsystems are considered as the equipment level, and key components such as valves, seals, and hydraulic components are considered as the component level, as shown in Table 1.

[0025] Table 1. Components of the Subsea Wellhead System

[0026] Based on the OREDA data book published by DNV and combined with field experience, this invention obtains the failure parameters of various components under typical operating conditions, and extracts the average repair time, failure rate and operating conditions of each type of component as input parameters for subsequent models.

[0027] Step 2: Reliability Analysis Based on Fuzzy Fault Tree and Markov Model Based on fault statistics, this invention establishes a polymorphic fuzzy fault tree model with "oil production tree failure" as the top event of the system, such as... Figure 1 As shown, the failure probability of the basic event is represented using an LR-type triangular fuzzy number, and its reference function is: ; Fuzzy logic gate operators (AND gate fuzzy operators / OR gate fuzzy operators) are applied to complete event propagation, calculating the failure rate of the top event and the probability importance of the bottom events. Through qualitative analysis of the subsea production tree fault tree, all possible factors leading to subsea production tree failure are listed, qualitatively identifying system weaknesses. Through quantitative analysis of the subsea production tree fault tree, the probabilities of bottom events are summarized, obtaining the reliability interval of the top event and the probability importance of the bottom events, quantitatively identifying system weaknesses.

[0028] For subsystems with obvious state transition relationships in their structure (such as hydraulic connectors), this invention establishes a state-space model of the subsea production tree system based on Markov modeling, such as... Figure 2 As shown, each state represents a different health level. Steady-state availability is solved using the transition probability matrix and state transition rate, and the sensitivity of key parameters is analyzed to quantitatively identify the bottlenecks in the system.

[0029] Based on the analysis and research of reliability theory, and considering the structural characteristics of the subsea Christmas tree system and the lack of statistical data and information on failures during operation, a fault tree model of the Christmas tree system was established for qualitative analysis. Triangular fuzzy numbers were applied to quantitatively analyze the reliability of the Christmas tree system, and fuzzy probability importance analysis was performed. The calculation results show that leakage, corrosion, blockage, seal failure, and malfunction (such as inability to open / close or lock / unlock) are relatively weak points in the Christmas tree system, and these five aspects should be regularly monitored and prevented.

[0030] By studying the failure mechanism of subsea production tree systems, a Markov model was established, and its reliability was quantitatively analyzed and calculated using field failure data of the subsea production tree system. A mathematical model of the subsea production tree system was established using the Markov model, and expressions for the system's instantaneous and steady-state availability were derived. Using statistical data from the OREDA datasheet as an example, the steady-state availability of the subsea production tree system was calculated to be 0.999671.

[0031] Step 3: Evaluation of network models based on probabilistic dependencies A probabilistic network model is constructed to model the faults of complex components, where nodes represent component states and edges represent conditional dependencies, such as... Figure 3 As shown, a conditional probability table is designed based on historical data, statistical experience, and expert knowledge. Considering fuzziness and polymorphism, uncertainty is encoded as a distribution range or confidence function. A time factor is introduced at the root node, and the node state is set to dynamically evolve, thus completing dynamic fuzzy modeling.

[0032] This model is used to characterize the impact of system structure dependencies on overall reliability, and does not involve fault diagnosis or state identification processes.

[0033] The fault tree model in step 2 is transformed into a BN model to establish a network model, resulting in a BN diagram of the system and subsystem branches. Quantitative analysis is then performed to complete the instance calculation of the Bayesian network model. Using Bayesian theoretical algorithms, the dynamic failure probability of different fault states of leaf nodes and the dynamic importance of root nodes are obtained. The results are compared with the FT calculation to verify the feasibility and accuracy of this invention in the reliability analysis of subsea oil production tree systems.

[0034] Step 4: Weighted fusion of information entropy from multiple model results To address the issues of differing evaluation focuses and inconsistent output ranges among the three types of model results obtained in steps 2 and 3, this invention proposes an uncertainty weight fusion algorithm based on information entropy. This algorithm weights and integrates the reliability results of the fuzzy fault tree, Markov model, and Bayesian network models to obtain the final comprehensive reliability index of the system. .

[0035] First, the system reliability assessment values ​​output by each model at analysis time t are collected and denoted as follows: , , Calculate the information entropy value based on the reliability interval length, probability distribution characteristics, or fuzzy membership function shape of each model's output. , , This serves as an indicator for measuring the uncertainty of the model's output.

[0036] The model fusion weights are constructed using the inverse of information entropy normalization, as shown in the following formula: ; Construct the final integrated reliability output: ; The outputs of the three models are combined into a decision matrix. ; The output of each model is normalized using the following formula: ; Calculate the information entropy of each model: ; Calculate entropy redundancy and adaptive weights, which can reflect the effective information content of the model and achieve adaptive adjustment.

[0037] ; ; For each sample i, reliability metrics are fused and a comprehensive reliability vector is finally calculated.

[0038] ; ; This method can adaptively adjust the influence weights of each model when the credibility of the results from multiple models is inconsistent, thereby improving the robustness and accuracy of the final reliability analysis results.

[0039] Example 2: The electronic device of the present invention includes a processor and a memory, wherein the memory stores computer-executable instructions, and the processor, when executing the instructions, implements the multi-model consistency reliability assessment method for an underwater vertical production tree system described in Embodiment 1.

[0040] Example 3: The computer-readable storage medium of the present invention stores a computer program, which, when executed by a processor, is used to implement the multi-model consistency reliability assessment method for an underwater vertical production tree system described in Embodiment 1.

[0041] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A multi-model consensus reliability assessment method for subsea vertical wellhead systems, characterized in that: The process includes system structure partitioning, multi-model reliability modeling, and model result fusion and evaluation. Specifically, the subsea vertical production tree system is structurally decomposed into a throttling module, a production tree body module, a hydraulic connector module, and a tubing suspension module. Based on reliability statistics and engineering experience parameters, a fuzzy fault tree model to describe the system's functional failure relationships, a state-space model to describe the state evolution characteristics of the system or subsystems, and a probabilistic network model to describe the structural dependencies of system components are established. Under the same evaluation conditions, the reliability evaluation results output by the fuzzy fault tree model, the state-space model, and the probabilistic network model are obtained and normalized. The corresponding information entropy value is calculated based on the uncertainty characteristics of each model's output results, and model fusion weights are constructed based on the information entropy. The multi-model reliability evaluation results are then weighted and integrated to obtain a comprehensive reliability evaluation index for the subsea vertical production tree system.

2. The multi-model consistency reliability assessment method for subsea vertical wellhead systems according to claim 1, characterized in that: The system structure is divided into system level, equipment level and component level. The system level is the subsea vertical Christmas tree system, the equipment level is the throttling module, the Christmas tree body module, the hydraulic connector module and the tubing suspension module, and the component level includes valves, sealing components, hydraulic joints and connecting components.

3. The multi-model consistency reliability assessment method for subsea vertical wellhead systems according to claim 1, characterized in that: The fuzzy fault tree model uses fuzzy set theory to describe the failure probability of basic events in intervals, and obtains the system reliability intervals and the importance ranking results of key components through fuzzy logic operations.

4. The multi-model consistency reliability assessment method for subsea vertical wellhead systems according to claim 1, characterized in that: The state-space model is a Markov state model, which calculates the steady-state availability of the system or subsystem under long-term operating conditions through state transition probabilities or state transition rate parameters.

5. The multi-model consistency reliability assessment method for subsea vertical wellhead systems according to claim 1, characterized in that: In the probabilistic network model, nodes represent the operating state of system components or subsystems, and the connections between nodes describe the dependencies at the structural or functional level. The probabilistic network model is used to perform reliability probability propagation calculations.

6. The multi-model consistency reliability assessment method for subsea vertical wellhead systems according to claim 1, characterized in that: The information entropy value is calculated based on the dispersion or uncertainty characteristics of the reliability assessment results output by each model. The smaller the information entropy value, the greater the weight obtained by the model in the fusion process.

7. The multi-model consistency reliability assessment method for subsea vertical wellhead systems according to claim 1, characterized in that: The comprehensive reliability assessment index is used to compare and analyze the reliability results of underwater vertical production tree systems under different structural configurations or different operating condition assumptions.

8. The multi-model consistency reliability assessment method for subsea vertical wellhead systems according to claim 1, characterized in that: The multi-model reliability assessment process does not involve system fault diagnosis, condition identification, or real-time risk prediction.

9. An electronic device, characterized in that: It includes a processor and a memory, wherein the memory stores computer-executable instructions, and the processor, when executing the instructions, implements the multi-model consistency reliability assessment method for subsea vertical production tree systems as described in any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a processor, is used to implement the multi-model consistency reliability assessment method for subsea vertical production tree systems as described in any one of claims 1 to 8.

Images