A maintenance sequence scheduling method for a marine oil underwater transportation system

Abstract

The present application belongs to the technical field of oil pipeline maintenance decision, and particularly relates to a maintenance sequence scheduling method for an offshore oil underwater transportation system. The scheduling method provides help for guaranteeing safe operation of the offshore oil underwater transportation system by dynamically evaluating state evolution of the offshore oil underwater transportation system and components thereof, and significantly reduces system downtime loss and operation and maintenance cost caused by fault propagation. The maintenance sequence scheduling method for the offshore oil underwater transportation system comprises the following steps: determining the structure of the offshore oil underwater transportation system and the relationship between components in the system, determining importance measurement indexes of the components, and dividing rolling time windows; calculating real-time comprehensive importance of the components; sorting the calculated real-time comprehensive importance of the components in descending order, and intervening in the maintenance sequence according to the sorting result; recording maintenance activities and updating the component state; and dividing the sequence according to the rolling time windows until all the time windows are traversed.

Description

Technical Field

[0001] This invention belongs to the field of oil pipeline maintenance and scheduling technology, and particularly relates to a maintenance sequence scheduling method for an underwater oil transportation system. Background Technology

[0002] Offshore subsea oil transportation systems, as a key component of subsea oil production, consist of pipelines, oil transfer stations, and auxiliary production facilities (see reference). Figure 1 As shown, oil transfer station L1 is responsible for initial receiving and transportation, transfer station L2 is responsible for intermediate and long-distance transportation, and auxiliary production facility L3 is responsible for final processing, metering, and distribution. The auxiliary production facility further includes refueling stations, metering stations, and pigging stations, depending on actual production requirements and pipeline layout needs. It is worth noting that, as a typical complex pipeline network with high interdependence, the offshore oil subsea transportation system exhibits strong coupling relationships in the structural connections and functional operation between its subsystems and components. Performance degradation or failure of any unit may lead to performance degradation or even cascading failures in other units, resulting in fault propagation.

[0003] Further research revealed that the core challenge of this fault propagation phenomenon lies in the significant dynamic evolution of propagation paths and failure impacts. Traditional fixed maintenance strategies struggle to perceive system state changes and fault propagation trends in real time, rendering them ineffective in addressing such dynamically evolving fault propagation processes. Furthermore, unlike land-based engineering, offshore oil and gas subsea transportation systems face more extreme operating environments, making offshore maintenance resource allocation difficult, and resulting in poor operability and high costs for seabed maintenance operations. This exacerbates the lag in response to rapidly changing fault propagation, severely impacting the safety and economic efficiency of offshore oil and gas subsea transportation systems. Summary of the Invention

[0004] This invention provides a maintenance sequence scheduling method for marine oil subsea transportation systems. This method effectively overcomes the limitations of traditional fixed maintenance strategies in that they cannot perceive system state changes and fault propagation trends in real time by dynamically evaluating the state evolution of marine oil subsea transportation systems and their components. It not only helps to ensure the safe and efficient operation of marine oil subsea transportation systems, but also significantly reduces system downtime losses and maintenance costs caused by fault propagation.

[0005] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0006] A maintenance sequence scheduling method for an underwater oil transport system includes the following steps:

[0007] Step S1: Determine the structure of the marine oil subsea transportation system and the relationships between its components;

[0008] The importance metrics for each component are determined based on its position, function, and impact on the system status within the marine oil and gas underwater transportation system.

[0009] Rolling time windows are defined based on the operational lifespan and general maintenance intervals of the marine oil and gas underwater transportation system.

[0010] Step S2: Calculate the real-time overall importance of each component;

[0011] Among them, the real-time comprehensive importance is jointly determined by three importance metrics: fault propagation importance, functional importance, and location importance.

[0012] Step S3: Sort the calculated real-time comprehensive importance of each component in descending order, and intervene in the maintenance sequence based on the sorting results;

[0013] Step S4: Record maintenance activities and update component status;

[0014] Step S5: Implement steps S2-S4 one by one according to the order of the rolling time window division, until all time windows are traversed.

[0015] Preferably, the process of calculating the real-time comprehensive importance of each component in step S2 specifically includes:

[0016] Step S21: Score the importance of fault propagation, functional importance, and location importance respectively;

[0017] Step S22: Use the entropy weight method to assign weights to the importance of fault propagation, functional importance, and location importance;

[0018] Step S23: Based on the fault propagation importance, functional importance, and location importance scores obtained in Step S21, and the fault propagation importance, functional importance, and location importance weights obtained in Step S22, calculate the real-time comprehensive importance of each component.

[0019] satisfy: Equation (11).

[0020] Preferably, the importance of fault propagation in step S21 is...

[0021] satisfy: Equation (5);

[0022] In equation (5), Let i be the total probability that a failure in component i will cause a failure in a downstream component, satisfying: Equation (3);

[0023] The total probability that it is triggered by an upstream component failure.

[0024] satisfy: Equation (4);

[0025] Let i be the set of downstream components of the i-th component. Let i be the set of upstream components of the i-th component; For dynamic propagation probability, satisfying: Equation (2);

[0026] In equation (2), satisfy: Equation (1);

[0027] In equation (1), It is the number of historical cases where the j-th component fails after the i-th component fails; It is the total number of historical faults of the i-th component; It is a coefficient adjusted based on expert experience and historical data, used to eliminate the influence of obviously irrelevant cases;

[0028] Functional importance, satisfying: Equation (6); where, This represents the loss of the i-th component at time l;

[0029] Location importance The specific topology is determined by the static topology of the marine oil and gas underwater transportation system and will not change with the real-time status of each component.

[0030] Preferably, the process of weighting the importance of fault propagation, functional importance, and location importance in step S22 specifically includes:

[0031] Step S221: Normalize the importance of fault propagation, functional importance, and location importance; wherein, the normalization process satisfies: Equation (7);

[0032] In equation (7), This represents the importance metric for the i-th component, where, ;

[0033] Step S222: Calculate the ratio of the value of the importance metric corresponding to each component to the sum of the values; wherein, under the same metric, the probability distribution ratio of each component in the interval [0,1] satisfies: Equation (8);

[0034] Step S223: Quantify the uncertainty information entropy of the importance metric.

[0035] get: Equation (9);

[0036] In equation (9), This is a normalization constant;

[0037] Step S224: Assign weights to the importance of fault propagation, functional importance, and location importance;

[0038] The weights of each importance metric satisfy the following: Equation (10);

[0039] In equation (10), The information utility value is expressed as follows: .

[0040] A more preferred approach also includes the following steps:

[0041] Step S6: Calculate the long-term expected cost of dynamic maintenance decisions and long-term system availability;

[0042] Among them, the long-term expected cost satisfies: Equation (16);

[0043] In equation (16), This indicates the current time window index. This represents the total number of time windows, satisfying: ; It is used to reflect the decline of future costs; This is the expected value;

[0044] Long-term system availability satisfies: Equation (17).

[0045] This invention provides a maintenance sequence scheduling method for an underwater oil and gas transportation system, comprising the following steps: Step S1: Determine the structure of the underwater oil and gas transportation system and the relationships between its components; determine the importance metrics of each component based on its position, function, and impact on the system's state; divide rolling time windows according to the system's operational lifespan and general maintenance intervals; Step S2: Calculate the real-time comprehensive importance of each component; Step S3: Sort the calculated real-time comprehensive importance of each component in descending order, and intervene in the maintenance sequence based on the sorting results; Step S4: Record maintenance activities and update component status; Step S5: Implement steps S2-S4 sequentially according to the rolling time window division order, until all time windows are traversed.

[0046] The maintenance sequence scheduling method for marine oil and gas underwater transportation systems, which features the above-described steps, has at least the following technical advantages compared to existing technologies:

[0047] (1) The maintenance sequence scheduling method for marine oil subsea transportation systems provided by this invention can realize dynamic evaluation of the status changes of marine oil subsea transportation systems and components; and by identifying key components, it achieves the purpose of sorting maintenance sequences according to the importance of components and maintenance requirements;

[0048] (2) The maintenance sequence scheduling method for marine oil and gas underwater transportation system provided by the present invention can promptly detect and handle critical faults, reduce system downtime and maintenance costs caused by fault propagation, and avoid over-maintenance, under-maintenance and maintenance delays, ultimately providing technical support for improving the production efficiency and safety of marine oil and gas underwater transportation system. Attached Figure Description

[0049] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the following drawings:

[0050] Figure 1 A reference schematic diagram showing the propagation paths of various component nodes in an offshore oil and gas underwater transportation system;

[0051] Figure 2 A flowchart illustrating a maintenance sequence scheduling method for an underwater oil transportation system provided by the present invention;

[0052] Figure 3 This is a schematic diagram of the comprehensive importance assessment results of each component calculated using the maintenance sequence scheduling method provided by the present invention;

[0053] Figure 4 This is a schematic diagram of the overall importance assessment results of each component calculated using a traditional fixed maintenance strategy without considering fault propagation;

[0054] Figure 5 A schematic diagram showing the overall importance assessment results of each component calculated using traditional fixed maintenance strategies while taking into account fault propagation. Detailed Implementation

[0055] This invention provides a maintenance sequence scheduling method for marine oil subsea transportation systems. This method effectively overcomes the limitations of traditional fixed maintenance strategies in that they cannot perceive system state changes and fault propagation trends in real time by dynamically evaluating the state evolution of marine oil subsea transportation systems and their components. It not only helps to ensure the safe and efficient operation of marine oil subsea transportation systems, but also significantly reduces system downtime losses and maintenance costs caused by fault propagation.

[0056] This invention provides a maintenance sequence scheduling method for an underwater oil transport system, such as... Figure 2 As shown, it includes the following steps:

[0057] Step S1: Determine the structure of the marine oil subsea transportation system and the relationships between its components. Based on the location, function, and impact on the system's state, determine the importance metrics for each component. Divide the system into rolling time windows based on its operational lifespan and general maintenance intervals.

[0058] It is worth noting that determining the structure of the marine oil subsea transportation system and the relationships between its components can provide information on the initial state of the system and its components, understand the degradation status of each component, and further determine the system's failure modes, thereby identifying the most appropriate importance metric.

[0059] Step S2: Calculate the real-time comprehensive importance of each component; wherein, the real-time comprehensive importance is determined by three importance metrics: fault propagation importance, functional importance, and location importance.

[0060] Based on completing step S1, step S2 is further implemented. In a preferred embodiment of the present invention, the process of calculating the real-time overall importance of each component in step S2 specifically includes:

[0061] Step S21: Score the importance of fault propagation, functional importance, and location importance respectively.

[0062] Specifically, the importance of this fault propagation satisfies: Equation (5);

[0063] In equation (5), Let i be the total probability that a failure in component i will cause a failure in a downstream component, satisfying: Equation (3); Let its total probability be triggered by an upstream component failure, satisfying: Equation (4).

[0064] Let i be the set of downstream components of the i-th component. Let i be the set of upstream components of the i-th component; For dynamic propagation probability, satisfying: Equation (2);

[0065] In equation (2), satisfy: Equation (1).

[0066] In equation (1), It is the number of historical cases where the j-th component fails after the i-th component fails; It is the total number of historical faults of the i-th component; It is a coefficient adjusted based on expert experience and historical data, used to eliminate the influence of obviously irrelevant cases.

[0067] It should be pointed out that, with Figure 1 Taking the offshore oil and gas subsea transportation system as an example, the actual propagation probability from component i to component j is determined by both the system topology and the current state of the components. When a component's state is close to failure, the probability of it generating cascading failures and being affected by other components increases. The worse the component's state, the higher its importance. Therefore, in terms of dynamic propagation probability, the state of component i at time t... In fact, within the range of [0,1], the closer the value is to 0, the better the component is; the closer the value is to 1, the more likely it is to fail.

[0068] Functional importance, satisfying: Equation (6); where, This represents the loss of the i-th component at time l.

[0069] Specifically, in equation (6) This refers to the degree of loss of core system functionality due to the current state of the i-th component, such as loss of downtime, throughput, and deliverability. Therefore, this functional importance metric is related to the component's state and changes over time. It's important to note that while the functional loss of each component also affects the functional loss of its downstream components, this primarily reflects its role in the propagation process. To avoid impacting upgrades due to faults, this analysis focuses only on the direct functional loss of components, excluding their indirect impact on subordinate systems.

[0070] And, location importance The specific topology is determined by the static topology of the marine oil and gas underwater transportation system and will not change with the real-time status of each component.

[0071] For example, a component's importance increases if it is the only connector in the system path. Conversely, its importance decreases if the component can be replaced and the corresponding path remains available. Location importance is calculated based on the minimum cut set derived from the fault tree, so only the fault tree needs to be calculated here.

[0072] Step S22: Use the entropy weight method to assign weights to the importance of fault propagation, functional importance, and location importance.

[0073] Specifically, in a preferred embodiment of the present invention, the process of weighting the importance of fault propagation, functional importance, and location importance in step S22 includes:

[0074] Step S221: Normalize the importance of fault propagation, functional importance, and location importance.

[0075] It is worth noting that for the three importance metrics of fault propagation importance, functional importance, and location importance, different importance metrics have different scales and value ranges; through this normalization process, these three importance metrics can be mapped to a unified dimensionless interval [0,1].

[0076] The normalization process satisfies the following: Equation (7);

[0077] In equation (7), This represents the importance metric for the i-th component, where, .

[0078] Step S222: Calculate the ratio of the value of the importance metric for each component to the sum of the values.

[0079] Specifically, for any metric, calculating the ratio of the value corresponding to each component to the sum of the values ​​ensures that data from different metrics are comparable. The greater the difference in the values ​​of importance metrics, the more uneven the distribution of the values; by normalizing the data into ratios to eliminate these differences, we can prepare for the subsequent step of calculating information entropy.

[0080] Among them, under the same index, the probability distribution ratio of each component in the interval [0,1] satisfies: Equation (8).

[0081] Step S223: Quantify the uncertainty information entropy of the importance metric.

[0082] get: Equation (9);

[0083] In equation (9), This is the normalization constant.

[0084] Step S224: Assign weights to the importance of fault propagation, functional importance, and location importance. The weights of each importance metric satisfy the following: Equation (10);

[0085] In equation (10), The information utility value is expressed as follows: .

[0086] It is worth noting that the purpose of this weight calculation is to assign a corresponding weight to each importance metric. During the propagation of the fault, The distribution of metrics can change significantly, for example, certain components may suddenly become key propagation nodes, thus requiring real-time reflection of changes in the amount of information in the metrics. Therefore, to avoid long-term decision bias in the initial weights due to data bias, adjustments are made in each subsequent time window. In all cases, it is necessary to recalculate and update the scores for fault propagation importance, functional importance, and location importance, as well as the weights for these scores.

[0087] Step S23: Based on the fault propagation importance, functional importance, and location importance scores obtained in Step S21, and the fault propagation importance, functional importance, and location importance weights obtained in Step S22, calculate the real-time comprehensive importance of each component.

[0088] satisfy: Equation (11).

[0089] It is worth noting that the importance of fault propagation and functional importance change dynamically with changes in component state and fault propagation process, leading to corresponding changes in their weights. Furthermore, even though location importance is related to system structure and does not change over time, its weight is closely related to the weights of other importance indicators, thus necessitating dynamic adjustment of its weights.

[0090] Step S3: Sort the calculated real-time comprehensive importance of each component in descending order, and intervene in the maintenance sequence based on the sorting results.

[0091] On the basis of completing step S2, step S3 is further implemented. It should be noted that in each time window, a new maintenance sequence can be generated by sorting the calculated real-time comprehensive importance of each component. This is set because of limited resources; it is assumed that only m components can be maintained within a time window, where 0 < m ≤ n. According to the maintenance sequence and maintenance resource constraints, it can be determined which components can be maintained within this time window, and corresponding maintenance strategies can be formulated according to the component status: such as major repair CM, minor repair PM, and non-repair NM. Among them, if the real-time comprehensive importance of multiple components is the same, components in a worse state can be given priority.

[0092] Step S4: Record the maintenance activities and update the component status.

[0093] On the basis of completing step S3, step S4 is further implemented. Specifically, after the maintenance is carried out, first record all maintenance activities and update the component status. Among them, the conversion of the component status follows the following model:

[0094] Equation (15);

[0095] In Equation (15), is the natural degradation rate of the component. The update of the component status will also synchronously update the propagation probability from component i to component j and the comprehensive importance in the next time window. According to the new component status and the evaluation result of the new comprehensive importance, it will also be reflected in the maintenance sequence of the next time window.

[0096] Step S5: Implement steps S2 - S4 one by one in the order of rolling time windows until all time windows are traversed.

[0097] Preferably, in a maintenance sequence scheduling method for an offshore oil underwater transportation system provided by the present invention, as Figure 2 shown, it further includes the following steps:

[0098] Step S6: Calculate the long-term expected cost and long-term system availability of the dynamic maintenance decision.

[0099] It should be noted that in the process of dynamic maintenance decision, the long-term expected cost is an important optimization goal, and it comprehensively evaluates the weighted average cost of future states through mathematical expectation. To calculate the long-term expected cost, first analyze the possible costs in the maintenance process as follows. Among them, the total cost includes direct maintenance costs, indirect downtime costs, penalty costs, etc. Further analysis shows that the maintenance cost is generally related to the maintenance strategy In contrast, downtime costs typically depend on the downtime caused by maintenance or malfunction. For example, if RM is selected, maintenance costs are higher; however, the downtime caused by maintenance is shorter, which actually reduces downtime costs.

[0100] The total cost for each specific time window satisfies: Equation (12).

[0101] In equation (12), For unit maintenance cost, It is the component's downtime within the current time window, which satisfies: Equation (13). This is the repair delay time. This represents the downtime caused by maintenance intervention, with both values ​​being 0 or ΔT. Further, assume that each time window allocates maintenance activities to at most m components. When a component malfunctions, it may cause downtime; however, due to resource constraints, the component cannot be repaired. In this case, the corresponding time window Δt will be recorded as the downtime of that component due to maintenance delay.

[0102] Based on this, the long-term expected cost is calculated.

[0103] satisfy: Equation (16);

[0104] In equation (16), This indicates the current time window index. This represents the total number of time windows, satisfying: ; It is used to reflect the decline of future costs; This is the expected value.

[0105] On the other hand, the long-term system availability of offshore oil and gas subsea transportation systems should be considered. Specifically, the system availability within a time window is determined by... This indicates that each simulation includes multiple time windows, the total time T, and each time window... z represents the current time window number (where, ), Represents the total number of time windows (where, No longer for each time window. To perform more detailed discretization, as long as a certain component i is within a certain time window If a component is unavailable (failed and not under repair, or under repair), then this time window period is recorded as the component's unavailability time. However, if this component returns to an available state, it will remain available in other time windows. After calculating the availability of a single component within the current time window, the path availability and system availability within this time window are then calculated to inform subsequent calculations of long-term availability. satisfy: Equation (14).

[0106] Then, long-term system availability is calculated. Specifically, long-term system availability satisfies: Equation (17).

[0107] Finally, to facilitate understanding of the present invention by those skilled in the art, based on Figure 1 The offshore subsea oil transport system shown implements the maintenance sequence scheduling process provided by this invention, and is illustrated below as an example. Specifically, this offshore subsea oil transport system consists of pipelines, oil transfer stations, transfer stations, and auxiliary production facilities, forming the backbone framework of the subsea transport network. Depending on actual production needs and pipeline distances, auxiliary production facilities may include various supporting facilities such as pressurization stations, metering stations, and pigging stations, which are not listed here. It should be noted that the structural and functional dependencies between components can lead to the risk of fault propagation, and may even exacerbate the severity of accidents. Furthermore, the cascading nature of fault propagation further increases the complexity of coordinating maintenance resources and effectively executing maintenance activities.

[0108] In this scenario, a three-layer network model is first constructed for the offshore oil subsea transportation system, with each layer containing multiple nodes. The first layer (L1) consists of the oil transfer station (responsible for crude oil reception and initial transmission), the second layer (L2) consists of the transfer station (responsible for transshipment and long-distance transportation), and the third layer (L3) consists of auxiliary production facilities (performing final processing, metering, and distribution). During actual construction, this offshore oil subsea transportation system will undergo natural degradation and require periodic maintenance. When cascading failures (CAFs) exist in the system, the consequences of accidents will be significantly amplified. Furthermore, this system contains at least two types of fault dependencies: firstly, inter-layer fault dependencies, existing between nodes in adjacent layers. For example, when the reliability of the source node in layer L1 or L2 falls below a threshold of 0.3, the fault will propagate along the transportation direction to connected nodes in subsequent layers. Secondly, intra-layer fault dependencies, existing between nodes within the same layer. When the reliability of the source node falls below 0.3, the fault will propagate between nodes within the same layer. Such propagation paths are determined by the configuration of the marine oil underwater transport system and expert experience.

[0109] Further statistical analysis reveals the following possible pathways within this marine oil underwater transport system: L1-1→L2-1→L3-1 (meaning: flow from node 1 of L1 to node 1 of L2, and flow from node 1 of L2 to node 1 of L3); L1-1→L2-1→L3-2; L1-1→L2-2→L3-1; L1-1→L2-2→L3-4; L1-1→L2-2→L3-5; L1-2→L2-2→L3-1; L1-2→L2-2→L3-4; L1-2→L2-2→L 3-5; L1-3→L2-3→L3-2; L1-3→L2-3→L3-5; L1-4→L2-3→L3-2; L1-4→L2-3→L3-5; L1-5→L2-3→L3-2; L1-5→L2-3→L3-5; L1-5→L2-4→L3-2; L1-5→L2-4→L3-3; L1-5→L2-4→L3-6; L1-6→L2-4→L3-2; L1-6→L2-4→L3-3; L1-6→L2-4→L3-6. The marine oil underwater transport system is considered to have failed only when all pathways fail. In other words, the marine oil underwater transport system can be considered a series-parallel system, and all pathways within the system can be considered to be in parallel. If there is a passable path in the system, it is considered usable.

[0110] Further input parameters are provided as shown below, and the maintenance sequence scheduling process provided by this invention is implemented. Table 1 shows the setting parameters for the degradation rate of the three-layer node components, Table 2 shows the setting parameters for maintenance activities, and Table 3 shows the calculation results of the basic propagation probability.

[0111] Table 1

[0112] ;

[0113] Table 2

[0114] ;

[0115] Table 3

[0116] ;

[0117] Based on the above data, the final evaluation results of the comprehensive importance of each component obtained using the maintenance sequence scheduling method provided by this invention (hereinafter referred to as Scenario 3) can be referenced as follows. Figure 3 As shown. Among them, the Figure 3 The horizontal axis represents time, and the vertical axis represents the overall importance of each component. Importance is represented by a value from 0 to 1; the higher the importance value, the more likely the component is to require maintenance.

[0118] In comparison, we further provide the overall importance assessment results calculated using the traditional fixed maintenance strategy without considering fault propagation (this scenario is labeled Scenario 1), and the overall importance assessment results calculated using the traditional fixed maintenance strategy but considering fault propagation (this scenario is labeled Scenario 2). For a schematic diagram of the overall importance assessment results of each component calculated using the traditional fixed maintenance strategy without considering fault propagation, please refer to [example diagram]. Figure 4 The diagram shown illustrates the overall importance assessment results for each component, calculated using traditional fixed maintenance strategies while considering fault propagation. (For reference...) Figure 5 As shown.

[0119] First, we compare the differences in the overall importance assessment results of each component using the traditional fixed maintenance strategy in Scenario 1 and Scenario 2, i.e., without considering fault propagation and with considering fault propagation. Specifically, without considering fault propagation (Scenario 1), the overall importance of each component fluctuates continuously; while considering fault propagation (Scenario 2), the overall importance of each component fluctuates significantly in the first 5 years, and then remains unchanged after the 7th year. The scenario without considering fault propagation and using the traditional fixed maintenance strategy only considers natural degradation. Because Scenario 1 does not consider accelerated degradation caused by fault propagation, its calculation results show that the expected degradation state of each component is slow, with only long-term continuous fluctuations. However, the calculation results considering fault propagation but still using the traditional fixed maintenance strategy show that the impact of fault propagation accelerates degradation, and the degradation of each component continues to accelerate; but the degradation effect gradually becomes ineffective between 5 and 7 years, and the overall importance of each component basically no longer changes afterward. Furthermore, it can be observed that when fault propagation is not considered, the fluctuation trends of the overall importance of components within each layer are similar; while after considering fault propagation, the fluctuation trends are not entirely the same. This is because some components are highly dependent on failures and are easily affected by failure propagation, resulting in larger fluctuations in their overall importance, such as component L3-2; while components with lower failure dependence and less susceptibility to failure propagation have smaller fluctuations in their overall importance, such as component L1-4. This, in turn, demonstrates the necessity of dynamic overall importance assessment.

[0120] Further comparison of Scenario 2 and Scenario 3 reveals differences in the overall importance assessment results calculated using the traditional fixed maintenance strategy while considering fault propagation, versus considering fault propagation and using the maintenance sequence scheduling method provided by this invention. It can be observed that although both scenarios consider the effect of fault propagation, the resulting fluctuations in overall importance differ significantly. Specifically, Figure 5Because the calculations used a traditional fixed maintenance strategy, the overall importance of each component fluctuated significantly, and fault propagation accelerated their degradation. However, after implementing the maintenance sequence scheduling method provided by this invention, the overall fluctuation in the overall importance of each component is relatively small. This is because this invention adjusts maintenance decisions and activities based on the priority of dynamic importance assessment results: by prioritizing the maintenance of more important components, their condition improves, and their importance decreases accordingly. Overall, the system implementing the maintenance sequence scheduling method provided by this invention will maintain a stable level of importance over the long term.

[0121] The comprehensive importance assessment results for the three scenarios above were summarized to obtain a schematic diagram of system and hierarchical state changes, as shown in Table 4 below.

[0122] Table 4

[0123] ;

[0124] Specifically, Table 4 visually illustrates the differences in system and hierarchical state changes across the three scenarios. The state is represented by a value between 0 and 1, with values ​​closer to 0 indicating a higher likelihood of failure and values ​​closer to 1 indicating a higher likelihood of being intact.

[0125] For scenarios one and two, it can be observed that, using the traditional fixed maintenance strategy, the system and hierarchical state without considering fault propagation is slightly better than the system and hierarchical state that does consider fault propagation. This indicates that when using the traditional fixed maintenance strategy to calculate and evaluate the performance of the submarine transmission system, the state assessment results are usually better than the actual operating conditions. However, if this phenomenon is ignored, maintenance personnel are prone to overestimating the system state and making incorrect maintenance decisions, resulting in under-maintenance and maintenance delays.

[0126] For scenarios two and three, it can be observed that, compared to using a traditional fixed maintenance strategy, the maintenance sequence scheduling method provided by this invention stabilizes the state at each level at around 0.5 starting from the 5th year, with the downward trend slowing down thereafter, and the system state still not dropping to 0 by the 10th year. The overall system state remains at state 1 for the first 6 years, with slight fluctuations in the last 4 years. This provides strong evidence for the performance improvement effect of the maintenance sequence scheduling method provided by this invention on the system.

[0127] It can be seen that the present invention provides a maintenance sequence scheduling method for marine oil and gas underwater transportation systems, which can realize dynamic evaluation as the system and component status changes, and identify key components, thereby avoiding over-maintenance, under-maintenance, and maintenance delays; and can effectively solve the maintenance decision-making problem of complex marine oil and gas extraction systems caused by maintenance resource constraints, thereby significantly improving the overall performance of the system.

[0128] This invention provides a maintenance sequence scheduling method for an underwater oil and gas transportation system, comprising the following steps: Step S1: Determine the structure of the underwater oil and gas transportation system and the relationships between its components; determine the importance metrics of each component based on its position, function, and impact on the system's state; divide rolling time windows according to the system's operational lifespan and general maintenance intervals; Step S2: Calculate the real-time comprehensive importance of each component; Step S3: Sort the calculated real-time comprehensive importance of each component in descending order, and intervene in the maintenance sequence based on the sorting results; Step S4: Record maintenance activities and update component status; Step S5: Implement steps S2-S4 sequentially according to the rolling time window division order, until all time windows are traversed.

[0129] The maintenance sequence scheduling method for marine oil and gas underwater transportation systems, which features the above-described steps, has at least the following technical advantages compared to existing technologies:

[0130] (1) The maintenance sequence scheduling method for marine oil and gas underwater transportation system provided by the present invention can realize the dynamic evaluation of the status changes of marine oil and gas underwater transportation system and components; and by identifying key components, it achieves the purpose of sorting the maintenance sequence according to the importance of the components and maintenance requirements.

[0131] (2) The maintenance sequence scheduling method for marine oil and gas underwater transportation system provided by the present invention can promptly detect and handle critical faults, reduce system downtime and maintenance costs caused by fault propagation, and avoid over-maintenance, under-maintenance and maintenance delays, ultimately providing technical support for improving the production efficiency and safety of marine oil and gas underwater transportation system.

[0132] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for scheduling maintenance sequences of a marine oil offshore transportation system, characterized in that, The method comprises the following steps: Step S1: determining the structure of the offshore oil underwater transportation system and the relationship between components in the system; According to the position, function and influence degree of each component on the system state in the offshore oil underwater transportation system, the importance measurement index of each component is determined; According to the operation life cycle and maintenance interval of the offshore oil underwater transportation system, a rolling time window is divided; Step S2: calculating the real-time comprehensive importance of each component; wherein the real-time comprehensive importance is determined by the fault propagation importance, the function importance and the position importance; Step S3: sorting the real-time comprehensive importance of each component in descending order, and intervening in the maintenance sequence according to the sorting result; Step S4: recording the maintenance activities and updating the component state; Step S5: implementing steps S2-S4 in sequence according to the rolling time window until all time windows are traversed; Fault propagation importance, satisfies: Equation (5); In formula (5), is the total probability of the downstream components failing when the i component fails, satisfying: Formula (3); The total probability for it to be triggered by an upstream component failure is satisfied by: Equation (4); a downstream component set of the ith component, an upstream component set of the ith component; a dynamic propagation probability, satisfying: Equation (2); In formula (2), satisfies: formula (1); In formula (1), is the number of historical cases of the jth component failure after the ith component failure; is the total number of historical failures of the ith component; is a coefficient modified according to expert experience and historical data, used to eliminate the influence of obviously irrelevant cases; Function importance, meet: Equation (6); wherein, represents the wear of the i component at time l; Positional importance Specifically determined by the static topology of the offshore oil underwater transportation system and does not change with the real-time state of each component.

2. The method of claim 1, wherein, The process of calculating the real-time comprehensive importance of each component in step S2 comprises: Step S21: scoring the fault propagation importance, the function importance and the position importance respectively; Step S22: using the entropy weight method to allocate weights to the fault propagation importance, the function importance and the position importance; Step S23: According to the fault propagation importance, the function importance, the location importance score obtained in step S21, the fault propagation importance, the function importance, the location importance weight obtained in step S22, the real-time comprehensive importance of each component is calculated, which satisfies: ; Equation (11).

3. The method of claim 2, wherein, The process of allocating weights to the fault propagation importance, the function importance and the position importance in step S22 comprises: Step S221: Normalization is performed on the fault propagation importance, the function importance, and the location importance; wherein the process of normalization satisfies: Equation (7); In formula (7), denotes an importance measure indicator of the i-th component, wherein, ; Step S222: calculating the ratio of the value of the importance measure index corresponding to each component to the sum of the values; wherein the probability distribution proportion of each component in the interval [0, 1] under the same index satisfies: Formula (8); Step S223: quantize the uncertainty information entropy of the importance measure indicator, to obtain: Equation (9); In formula (9), is a normalization constant; Step S224: allocating weights to the fault propagation importance, the function importance and the position importance; wherein the weights of each importance metric indicator satisfy: Equation (10); In formula (10), represents the information utility value, satisfying: .

4. The method of claim 1, wherein, The method further comprises the following steps: Step S6: calculating the long-term expected cost of the dynamic maintenance decision and the long-term system availability; wherein the long-term expected cost satisfies: Equation (16); In formula (16), denotes the current time window index, denotes the total number of time windows, satisfying: ; for reflecting the decay of the future cost; is the expected value; Long term system availability, meet: Equation (17).

Images