Fault prediction method and system for FPSO riser support system

Abstract

The invention discloses a fault prediction method and system for an FPSO riser support system, and relates to the technical field of fault prediction, and the method comprises the steps: obtaining current fault data related to the fault state of the FPSO riser support system, and obtaining a final fault type through combining with historical fault data; obtaining historical data corresponding to each final fault type to calculate a maintenance importance coefficient, a pollution coefficient and a yield reduction coefficient; calculating a key coefficient of each fault type according to the maintenance importance coefficient, the pollution coefficient and the output reduction coefficient, and judging the maintenance urgency of the fault type of each FPSO riser support system according to the key coefficient; in this way, the fault of the FPSO riser support system can be predicted in a multi-level manner, and the accuracy of fault prediction is ensured; and under the condition that a plurality of faults possibly occur at the same time, which fault is the most emergency and critical fault can be accurately judged, and resources are reasonably allocated for preferential maintenance and processing.

Description

Technical Field

[0001] The present invention relates to the technical field of fault prediction, and in particular to a fault prediction method and system for an FPSO riser support system. Background Art

[0002] The fault prediction method of the FPSO (Floating Production Storage and Offloading) riser support system usually refers to predicting possible failures or damages of the riser support system through real-time monitoring, data analysis and machine learning. These systems are subjected to extreme marine conditions in offshore operating environments, including wind and waves, temperature changes, and seawater corrosion. Therefore, fault prediction can be achieved by monitoring the health of key components of the riser support system (such as support structures, connection points, sensors, etc.), collecting vibration, stress, displacement and other data, and analyzing potential risk points. When an abnormal pattern is detected, the system will issue an early warning in a timely manner, thereby providing decision support for maintenance personnel and avoiding potential equipment failures or safety accidents.

[0003] However, the above prediction methods often only predict the failure of the FPSO riser support system, and only consider the failure of a single system or component. The failure of the FPSO riser support system may be caused by multiple factors. A single fault prediction may lead to low fault prediction accuracy. In addition, when multiple faults may occur at the same time, how to accurately judge which fault is the most urgent and critical, and how to allocate resources for priority repair and processing are also problems that have not been fully solved in current prediction methods. Summary of the invention

[0004] The purpose of the present invention is to solve the above-mentioned problems and to provide a fault prediction method and system for a FPSO riser support system.

[0005] In a first aspect of the present invention, a method for predicting a fault of a FPSO riser support system is first proposed, the method comprising: Acquire data on the fault state of the FPSO riser support system as current fault data, calculate similarity with each type of historical fault data of the FPSO riser support system, and predict the fault type of the FPSO riser support system as a final fault type; For each final fault type, obtain the historical maintenance data corresponding to each fault type, and calculate the maintenance importance coefficient based on the historical maintenance data; Obtain historical pollution data corresponding to each fault type, and calculate the pollution coefficient based on the historical pollution data; Obtain the historical production reduction data corresponding to each fault type, and calculate the production reduction coefficient based on the historical maintenance data; The key coefficient of each fault type is calculated according to the maintenance importance coefficient, pollution coefficient and production reduction coefficient, and the maintenance urgency of each FPSO riser support system fault type is determined according to the key coefficient.

[0006] Optionally, calculating the similarity with each historical fault data of the FPSO riser support system and predicting the fault type of the FPSO riser support system as the final fault type includes: Each type of historical fault data is recorded as target historical fault data, and the current fault data and the target historical fault data are preprocessed; Construct a distance matrix, where each element is the distance between the current fault data point and the target historical fault data point; A dynamic programming algorithm is used to calculate the shortest path in the distance matrix to find the best match between the current fault data sequence and the target historical fault data sequence; According to the calculated shortest path, align the current fault data sequence with the target historical fault data sequence; According to the aligned current fault data sequence and the target historical fault data sequence, the similarity between them is calculated; The calculated similarity is compared with the similarity threshold. If the similarity is greater than the similarity threshold, the corresponding historical fault type is used as the predicted fault type. All predicted fault types are taken as the final fault types.

[0007] Optionally, calculating the maintenance importance coefficient based on historical maintenance data includes: For each final fault type, obtain the number of times the fault type occurs in the historical data, obtain the maintenance time spent on each fault, and calculate the mean maintenance time as the maintenance time coefficient; Obtain the average maintenance time interval, and obtain the time interval between the current time and the last maintenance of the fault, and subtract the time between the current time and the last maintenance from the average maintenance time interval to obtain the maintenance interval coefficient; The maintenance importance coefficient is calculated based on the maintenance time coefficient and the maintenance interval coefficient. The calculation formula is: ; In the formula, is the maintenance importance factor, are the maintenance time coefficient, maintenance interval coefficient and the total number of occurrences of the fault type, They are The preset scaling factor of Both are greater than 0.

[0008] Optionally, calculating the pollution coefficient based on historical pollution data includes: For each final fault type, obtain the number of occurrences of this fault type in the historical data, and obtain the volume of seawater discharged from the oil storage each time a fault occurs. The volume of seawater discharged from the oil storage each time a fault occurs is calibrated as , The sequential number representing the volume of the oil storage discharged into the seawater each time a failure occurs, and , is the total number of faults that occurred, and is a positive integer; Calculate the average volume of oil discharged into the seawater , the calculation formula is: ; Calculate the contamination factor , the calculation formula is: , In the formula, is the total number of occurrences of the fault type, They are The preset scaling factor of Both are greater than 0.

[0009] Optionally, obtaining historical production reduction data corresponding to each fault type, and calculating the production reduction coefficient according to the historical maintenance data includes: For each final fault type, the number of occurrences of the fault type in the historical data is obtained, and the difference between the preset production oil volume and the actual production oil volume when each fault occurs is obtained as the reduced production oil volume of that time; Add up all the reduced oil production when each failure occurs to obtain the reduction coefficient.

[0010] Optionally, calculating a critical coefficient of each fault type according to the maintenance importance coefficient, the pollution coefficient and the production reduction coefficient, and judging the maintenance urgency of each FPSO riser support system fault type according to the critical coefficient includes: ; In the formula, is the critical coefficient of the fault type, They are maintenance importance coefficient, pollution coefficient and production reduction coefficient; They are The preset scaling factor of All are greater than 0; According to the size of the critical coefficient, the fault types of each FPSO riser support system are ranked from high to low in terms of maintenance urgency.

[0011] In a second aspect of the present invention, a fault prediction system for a FPSO riser support system is provided, the system comprising: Final fault type module: obtains data on the fault status of the FPSO riser support system as current fault data, calculates the similarity with each type of historical fault data of the FPSO riser support system, and predicts the fault type of the FPSO riser support system as the final fault type; Maintenance importance module: For each final fault type, obtain the historical maintenance data corresponding to each fault type, and calculate the maintenance importance coefficient based on the historical maintenance data; Pollution module: obtains historical pollution data corresponding to each fault type and calculates the pollution coefficient based on the historical pollution data; Production reduction module: obtains historical production reduction data corresponding to each fault type, and calculates the production reduction coefficient based on historical maintenance data; Maintenance module: Calculate the key coefficient of each fault type based on the maintenance importance coefficient, pollution coefficient and production reduction coefficient, and judge the maintenance urgency of the fault type of each FPSO riser support system based on the key coefficient.

[0012] Beneficial effects of the present invention: The present invention proposes a fault prediction method and system for a FPSO riser support system. The method obtains current fault data on the fault state of the FPSO riser support system and calculates the similarity with each type of historical fault data of the FPSO riser support system to predict the fault type of the FPSO riser support system as the final fault type; for each final fault type, obtains historical maintenance data corresponding to each fault type to calculate the maintenance importance coefficient; obtains historical pollution data corresponding to each fault type to calculate the pollution coefficient; obtains historical production reduction data corresponding to each fault type to calculate the production reduction coefficient; calculates the key coefficient of each fault type according to the maintenance importance coefficient, the pollution coefficient and the production reduction coefficient, and judges the maintenance urgency of the fault type of each FPSO riser support system according to the key coefficient; in this way, the faults of the FPSO riser support system can be predicted at multiple levels to ensure the accuracy of the fault prediction; and in the case where multiple faults may occur at the same time, it can accurately judge which fault is the most urgent and critical, and reasonably allocate resources for priority maintenance and processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The present invention will be further described below in conjunction with the accompanying drawings.

[0014] Figure 1 A flow chart of a fault prediction method for a FPSO riser support system; Figure 2 A framework diagram of a fault prediction system for a FPSO riser support system. DETAILED DESCRIPTION

[0015] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0016] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in the field without making any creative work shall fall within the scope of protection of the present invention.

[0017] The embodiment of the present invention provides a method for predicting failure of a FPSO riser support system. Figure 1 , Figure 1 A flowchart of a method for predicting a fault of a FPSO riser support system provided by an embodiment of the present invention. The method comprises the following steps: Acquire data on the fault state of the FPSO riser support system as current fault data, calculate similarity with each type of historical fault data of the FPSO riser support system, and predict the fault type of the FPSO riser support system as a final fault type; For each final fault type, obtain the historical maintenance data corresponding to each fault type, and calculate the maintenance importance coefficient based on the historical maintenance data; Obtain historical pollution data corresponding to each fault type, and calculate the pollution coefficient based on the historical pollution data; Obtain the historical production reduction data corresponding to each fault type, and calculate the production reduction coefficient based on the historical maintenance data; The key coefficient of each fault type is calculated according to the maintenance importance coefficient, pollution coefficient and production reduction coefficient, and the maintenance urgency of each FPSO riser support system fault type is determined according to the key coefficient.

[0018] Based on a fault prediction method for a FPSO riser support system provided in an embodiment of the present invention, the above-mentioned method can predict the faults of the FPSO riser support system at multiple levels to ensure the accuracy of the fault prediction; and in the case where multiple faults may occur at the same time, it can accurately determine which fault is the most urgent and critical, and reasonably allocate resources for priority repair and processing.

[0019] In one embodiment, calculating the similarity with each historical fault data of the FPSO riser support system and predicting the fault type of the FPSO riser support system as the final fault type includes: Each type of historical fault data is recorded as target historical fault data, and the current fault data and the target historical fault data are preprocessed; Construct a distance matrix, where each element is the distance between the current fault data point and the target historical fault data point; A dynamic programming algorithm is used to calculate the shortest path in the distance matrix to find the best match between the current fault data sequence and the target historical fault data sequence; According to the calculated shortest path, align the current fault data sequence with the target historical fault data sequence; According to the aligned current fault data sequence and the target historical fault data sequence, the similarity between them is calculated; The calculated similarity is compared with the similarity threshold. If the similarity is greater than the similarity threshold, the corresponding historical fault type is used as the predicted fault type. All predicted fault types are taken as the final fault types.

[0020] It should be noted that the similarity threshold is set by professionals based on actual circumstances and will not be limited or elaborated on in detail.

[0021] It should be noted that the fault status data in the FPSO riser support system can include various key parameters, such as vibration, stress, displacement, temperature, corrosion level, load change, current data, etc., which can reflect the operating health of the riser support system. These data can be collected in real time through sensors and monitoring equipment installed on various components of the system. These sensors usually include strain gauges, accelerometers, temperature sensors, pressure sensors, vibration sensors, etc. The fault-related data monitored in real time is transmitted to the central processing system through the data acquisition system for analysis and prediction. Through regular or real-time data analysis, potential faults can be discovered in time to ensure the safe and reliable operation of the FPSO riser support system.

[0022] It should be noted that the purpose of data preprocessing is to convert the data into a form suitable for calculating similarity. The preprocessing steps include: data cleaning: removing outliers, noise data, etc.; standardization or normalization: unifying the scale of the data to avoid the adverse effects of differences between different dimensions on the calculation results; time series alignment: ensuring that the time windows of the current fault data and the target historical fault data are consistent, and it may be necessary to perform time interpolation or resampling of the historical data; predicting and determining the fault type of the FPSO riser support system based on the preprocessed current fault data and the target historical fault data.

[0023] In one implementation, determining the potential fault types of the FPSO riser support system by the above method has significant benefits. First, it can accurately predict the potential fault types based on the similarity between historical fault data and real-time monitoring data, rather than relying solely on experience or preset fault models, thereby greatly improving the accuracy and foresight of the prediction; second, this method can identify subtle changes and evolution trends in fault modes, helping maintenance personnel to promptly discover hidden faults and potential risks, and avoid major safety accidents and equipment damage caused by failure to detect faults in a timely manner; in addition, this method can predict the fault types of multiple FPSO riser support systems. Compared with the prediction of a single fault type, this method can simultaneously identify multiple potential fault types, thereby providing a more comprehensive fault warning, reducing the risk caused by missed diagnosis of a single fault prediction, and ensuring that maintenance decisions are more accurate and comprehensive.

[0024] In one implementation, the application of a fault prediction method based on the dynamic time warping (DTW) algorithm in the FPSO riser support system has significant advantages; first, by accurately matching current fault data with historical fault data, it can identify potential risks similar to historical fault patterns, thereby providing early warnings and helping maintenance personnel to promptly detect possible fault types. This method does not rely on simple threshold judgments, but instead calculates the shortest path and similarity, taking into account the time series characteristics of fault evolution, and can cope with complex and nonlinear fault patterns. Second, the dynamic programming algorithm enables the system to handle situations where the time series length is inconsistent or there is data noise, enhancing the robustness and accuracy of the prediction.

[0025] In one embodiment, calculating the maintenance importance coefficient based on historical maintenance data includes: For each final fault type, obtain the number of times the fault type occurs in the historical data, obtain the maintenance time spent on each fault, and calculate the mean maintenance time as the maintenance time coefficient; Obtain the average maintenance time interval, and obtain the time interval between the current time and the last maintenance of the fault, and subtract the time between the current time and the last maintenance from the average maintenance time interval to obtain the maintenance interval coefficient; The maintenance importance coefficient is calculated based on the maintenance time coefficient and the maintenance interval coefficient. The calculation formula is: ; In the formula, is the maintenance importance factor, are the maintenance time coefficient, maintenance interval coefficient and the total number of occurrences of the fault type, They are The preset scaling factor of Both are greater than 0.

[0026] It should be noted that It is set by professionals according to the actual situation. Generally, The sum is 1, and no specific limitation or elaboration is given; It should be noted that the data involved in the above calculations can be obtained in a variety of ways. First, the number of faults and the maintenance time spent on each repair are usually derived from the system's maintenance records and fault logs, which are generated and archived in real time by field engineers or automated monitoring systems. The maintenance time can be obtained through work orders or maintenance reports recorded by maintenance personnel. In addition, the average time interval for maintenance can be statistically analyzed through historical maintenance data and calculated based on the repair time interval after each system failure. For the time interval between the current time of the fault and the last repair, it is necessary to capture and record the timestamp of each fault in real time through the fault monitoring system. All of this data is usually collected through sensors, maintenance management systems (such as CMMS, EAM systems) or automated monitoring systems, and stored in the company's database for subsequent analysis and calculation.

[0027] It should be noted that the larger the maintenance importance coefficient of each final fault type, the more urgent and critical the corresponding fault type FPSO riser support system fault is, and the more priority maintenance and processing is needed. Because the larger the maintenance importance coefficient of each final fault type, it means that the fault type has a higher frequency of occurrence in historical maintenance data, the time required for maintenance is longer, and the maintenance interval is shorter, indicating that this type of fault not only occurs frequently, but also has a greater impact on the system each time it occurs, and the maintenance difficulty is higher. If not handled in time, it may cause the system to shut down for a long time or cause more serious damage. In addition, a large value of the maintenance interval coefficient may also indicate that the fault has a high recurrence and accumulation. Once it is not repaired in time, it may aggravate the loss of equipment or extend the maintenance time. Therefore, the larger the maintenance importance coefficient, the higher the potential risk of this fault type in system operation, and it is necessary to prioritize the allocation of maintenance resources and take measures as soon as possible to avoid more serious failures or shutdowns in the system, so as to ensure the safety and stability of the system.

[0028] In one implementation, analyzing the maintenance importance coefficient has the following benefits for determining which fault is the most urgent and critical when multiple faults of the FPSO riser support system may occur simultaneously, and how to allocate resources for priority maintenance and processing: The maintenance importance factor can provide a quantitative standard to help decision makers accurately identify the most urgent and critical types of faults when judging that multiple faults may occur simultaneously. By comprehensively considering the frequency of faults, maintenance time and maintenance interval, the maintenance importance factor gives each fault type a clear priority. In the case of multiple concurrent faults, faults with higher maintenance importance factors usually mean that they have a greater impact on the system, a longer recovery time or a higher probability of occurrence. Therefore, giving priority to repairing these faults can minimize potential safety risks and economic losses. Such prioritization can not only avoid minor faults from being ignored due to incorrect processing order, but also ensure that limited maintenance resources are used most effectively. By scientifically allocating resources, maintenance personnel can focus on solving the most critical fault problems, ensuring the stability and reliability of the FPSO riser support system, while improving overall operational efficiency and reducing downtime and maintenance costs.

[0029] In one embodiment, calculating the pollution coefficient based on historical pollution data includes: For each final fault type, obtain the number of occurrences of this fault type in the historical data, and obtain the volume of seawater discharged from the oil storage each time a fault occurs. The volume of seawater discharged from the oil storage each time a fault occurs is calibrated as , The sequential number representing the volume of the oil storage discharged into the seawater each time a failure occurs, and , is the total number of faults that occurred, and is a positive integer; Calculate the average volume of oil discharged into the seawater , the calculation formula is: ; Calculate the contamination factor , the calculation formula is: , In the formula, is the total number of occurrences of the fault type, They are The preset scaling factor of Both are greater than 0.

[0030] It should be noted that It is set by professionals according to the actual situation. Generally, 1, no specific limitation or elaboration is given; It should be noted that the number of historical failures and the volume of seawater discharged from the oil storage can be obtained through a maintenance management system (such as a CMMS or EAM system) or an accident report database.

[0031] It should be noted that the larger the pollution coefficient of each final fault type, the more urgent and critical the corresponding fault type of FPSO riser support system fault is, and the more priority it needs to be repaired and handled, because the larger the pollution coefficient of each final fault type, the more it means that once this fault type occurs, a large amount of stored oil will be discharged into the seawater, causing serious environmental pollution. This will not only have a great negative impact on the marine ecosystem, but may also trigger penalties at the legal and regulatory levels and strong reactions from public opinion, affecting the company's reputation and long-term operations. In addition, pollution accidents usually require more resources and time for emergency treatment and cleanup, adding additional economic burdens and risks to the company. Therefore, fault types with larger pollution coefficients are often proportional to the severity of environmental damage. Timely repair of these faults can not only effectively avoid the occurrence of pollution accidents and reduce damage to the ecological environment, but also minimize the resulting economic losses and legal risks. For this reason, fault types with higher pollution coefficients should be regarded as the most urgent and critical fault types, and should be repaired and handled first, so as to ensure the stable operation of the system, protect the environment, and at the same time improve the company's sense of social responsibility and sustainable development capabilities.

[0032] In one implementation, the benefits of analyzing the pollution coefficient for determining which fault is the most urgent and critical when multiple faults of the FPSO riser support system may occur simultaneously, and how to allocate resources for priority repair and processing are as follows: when multiple faults may occur simultaneously, the pollution coefficient can provide decision makers with a clear priority ranking, helping to accurately determine which fault types are the most urgent and critical. Faults with higher pollution coefficients usually mean that they pose a greater potential threat to the environment, which may cause serious ecological pollution, environmental disasters and their chain reactions, such as legal proceedings, economic losses and damage to the public image faced by the company. Therefore, using the pollution index to assess the environmental impact of faults can help the maintenance team focus resources on the faults that need to be handled immediately, avoiding irreparable losses caused by ignoring faults with high pollution risks. At the same time, this method helps to reasonably allocate limited maintenance resources, ensure that the faults with the greatest environmental risks are repaired first, prevent catastrophic consequences, and thus improve the emergency response efficiency and overall operational sustainability of the FPSO riser support system. Through such a resource allocation strategy, we can not only protect the marine environment and fulfill corporate social responsibility, but also minimize the negative impacts and potential high costs caused by environmental pollution, and ensure the stability and long-term healthy operation of the system.

[0033] In one embodiment, obtaining historical production reduction data corresponding to each fault type and calculating the production reduction coefficient according to the historical maintenance data includes: For each final fault type, the number of occurrences of the fault type in the historical data is obtained, and the difference between the preset production oil volume and the actual production oil volume when each fault occurs is obtained as the reduced production oil volume of that time; Add up all the reduced oil production when each failure occurs to obtain the reduction coefficient.

[0034] It should be noted that the above-mentioned preset oil production volume is set by professionals according to actual conditions and will not be limited or elaborated on in detail.

[0035] It should be noted that the actual oil production is collected by the real-time production monitoring system or oil and gas metering equipment on the FPSO platform. Equipment such as flow meters, pressure sensors and temperature sensors can record production changes in real time. These monitoring equipment and data acquisition systems will automatically record the difference between the actual oil production and the preset target when each failure occurs. The occurrence of historical failures and related production reduction data can be queried through the failure reports and maintenance records in the production management system (such as SCADA system) or maintenance management system (such as CMMS).

[0036] It should be noted that the larger the reduction coefficient of each final fault type, the more urgent and critical the corresponding fault type is among the faults of the FPSO riser support system, and the more priority it needs to be repaired and handled. Because the larger the reduction coefficient of each final fault type, the more serious the impact of the fault on the production capacity of the FPSO riser support system, and the greater the production loss. In offshore oil production, production is directly related to the economic benefits of the enterprise. Therefore, any decline in production caused by a fault may have a significant impact on profitability and production targets. When the reduction coefficient of a fault type is large, it means that the production capacity is greatly reduced when the fault occurs, which may lead to a long-term shutdown or reduced efficiency, and thus cause greater economic losses and safety hazards. Therefore, giving priority to repairing and handling these faults with large reduction coefficients can not only resume production in time and reduce losses, but also ensure the operational stability and reliability of the FPSO riser support system and ensure the smooth operation of the entire production chain. This priority allocation method can help resources be used most effectively and reduce the comprehensive risks brought by faults.

[0037] In one implementation, the benefits of analyzing the reduction coefficient for determining which fault is the most urgent and critical when multiple faults of the FPSO riser support system may occur simultaneously, and how to allocate resources for priority repair and processing are as follows: The reduction factor plays an important role in determining the situation where multiple faults may occur simultaneously in the FPSO riser support system. When multiple faults occur simultaneously, the reduction factor can provide a quantitative basis to help prioritize the identification of those faults that have the most serious impact on production. By calculating the reduction factor for each type of fault, it is clear which faults cause the greatest production loss, so that resources can be focused on these key faults for repair and avoid further decline in production efficiency. This not only ensures that production is restored as soon as possible, but also optimizes the allocation of maintenance resources, reduces downtime and maintenance costs, and improves the operational efficiency of the entire system. Through this mechanism, the rational allocation of resources can maximize benefits and ensure that the most pressing production problems are solved in a timely manner in emergency situations.

[0038] In one embodiment, calculating the critical coefficient of each fault type according to the maintenance importance coefficient, the pollution coefficient and the production reduction coefficient, and judging the maintenance urgency of each FPSO riser support system fault type according to the critical coefficient includes: ; In the formula, is the critical coefficient of the fault type, They are maintenance importance coefficient, pollution coefficient and production reduction coefficient; They are The preset scaling factor of All are greater than 0; According to the size of the critical coefficient, the fault types of each FPSO riser support system are ranked from high to low in terms of maintenance urgency.

[0039] It should be noted that It is set by professionals according to the actual situation. Generally, The sum of is 1, for example They can be 0.4, 0.3, 0.3, or other numbers, and are not specifically limited. In addition, commonly used normalization methods include Min-Max normalization, Z-Score normalization, etc. The specific method is selected by professionals according to the actual situation, and is not specifically limited or elaborated. In one implementation, the key coefficient of each fault type is calculated based on the maintenance importance coefficient, pollution coefficient and production reduction coefficient, and the maintenance urgency is judged based on the key coefficient, which can help accurately identify the most urgent and critical fault types. By sorting various types of faults and giving priority to those faults that have a greater impact on production and the environment, the potential risks and losses caused by the faults can be minimized. First, the maintenance importance coefficient helps to assess the difficulty of repair and historical impact of the fault, the pollution coefficient reflects the potential threat of the fault to the environment, and the production reduction coefficient intuitively reveals the size of the production loss. Taking these factors into consideration can help to comprehensively evaluate the urgency of different faults, ensure that resources can be invested in the most urgent problems, avoid wasting maintenance resources, and minimize system downtime and production interruptions. Ultimately, this method can improve the efficiency and accuracy of the entire maintenance decision, optimize the maintenance and management of the FPSO riser support system, and ensure the long-term stable operation of the equipment.

[0040] Based on the same inventive concept, the present invention also provides a fault prediction system for a FPSO riser support system. Figure 2 , Figure 2 A framework diagram of a fault prediction system for a FPSO riser support system provided by an embodiment of the present invention, the system comprising: Final fault type module: obtains data on the fault status of the FPSO riser support system as current fault data, calculates the similarity with each type of historical fault data of the FPSO riser support system, and predicts the fault type of the FPSO riser support system as the final fault type; Maintenance importance module: For each final fault type, obtain the historical maintenance data corresponding to each fault type, and calculate the maintenance importance coefficient based on the historical maintenance data; Pollution module: obtains historical pollution data corresponding to each fault type and calculates the pollution coefficient based on the historical pollution data; Production reduction module: obtains historical production reduction data corresponding to each fault type, and calculates the production reduction coefficient based on historical maintenance data; Maintenance module: Calculate the key coefficient of each fault type based on the maintenance importance coefficient, pollution coefficient and production reduction coefficient, and judge the maintenance urgency of the fault type of each FPSO riser support system based on the key coefficient.

[0041] A fault prediction system for a FPSO riser support system provided in an embodiment of the present invention can predict faults of the FPSO riser support system at multiple levels through the above-mentioned method, thereby ensuring the accuracy of fault prediction; and in the case where multiple faults may occur at the same time, it can accurately determine which fault is the most urgent and critical, and reasonably allocate resources for priority repair and processing.

[0042] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be used to artificially limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A method for predicting a fault of a FPSO riser support system, characterized in that: The following steps are involved: Acquire data on the fault state of the FPSO riser support system as current fault data, calculate similarity with each type of historical fault data of the FPSO riser support system, and predict the fault type of the FPSO riser support system as a final fault type; For each final fault type, obtain the historical maintenance data corresponding to each fault type, and calculate the maintenance importance coefficient based on the historical maintenance data; Obtain historical pollution data corresponding to each fault type, and calculate the pollution coefficient based on the historical pollution data; Obtain the historical production reduction data corresponding to each fault type, and calculate the production reduction coefficient based on the historical maintenance data; The key coefficient of each fault type is calculated according to the maintenance importance coefficient, pollution coefficient and production reduction coefficient, and the maintenance urgency of each FPSO riser support system fault type is determined according to the key coefficient.

2. A fault prediction method for a FPSO riser support system according to claim 1, characterized in that: The similarity between each historical fault data of the FPSO riser support system is calculated, and the fault type of the FPSO riser support system is predicted as the final fault type including: Each type of historical fault data is recorded as target historical fault data, and the current fault data and the target historical fault data are preprocessed; Construct a distance matrix, where each element is the distance between the current fault data point and the target historical fault data point; A dynamic programming algorithm is used to calculate the shortest path in the distance matrix to find the best match between the current fault data sequence and the target historical fault data sequence; According to the calculated shortest path, align the current fault data sequence with the target historical fault data sequence; According to the aligned current fault data sequence and the target historical fault data sequence, the similarity between them is calculated; The calculated similarity is compared with the similarity threshold. If the similarity is greater than the similarity threshold, the corresponding historical fault type is used as the predicted fault type. All predicted fault types are taken as the final fault types.

3. A fault prediction method for a FPSO riser support system according to claim 1, characterized in that: Calculation of maintenance importance factors based on historical maintenance data includes: For each final fault type, obtain the number of times the fault type occurs in the historical data, obtain the maintenance time spent on each fault, and calculate the mean maintenance time as the maintenance time coefficient; Obtain the average maintenance time interval, and obtain the time interval between the current time and the last maintenance of the fault, and subtract the time between the current time and the last maintenance from the average maintenance time interval to obtain the maintenance interval coefficient; The maintenance importance coefficient is calculated based on the maintenance time coefficient and the maintenance interval coefficient. The calculation formula is: ； In the formula, is the maintenance importance factor, are the maintenance time coefficient, maintenance interval coefficient and the total number of occurrences of the fault type, They are The preset scaling factor of Both are greater than 0.

4. A method for predicting a fault of a FPSO riser support system according to claim 1, characterized in that: Calculation of pollution coefficients based on historical pollution data includes: For each final fault type, obtain the number of occurrences of this fault type in the historical data, and obtain the volume of seawater discharged from the oil storage each time a fault occurs. The volume of seawater discharged from the oil storage each time a fault occurs is calibrated as , The sequential number representing the volume of the oil storage discharged into the seawater each time a failure occurs, and , is the total number of faults that occurred, and is a positive integer; Calculate the average volume of oil discharged into the seawater , the calculation formula is: ; Calculate the contamination factor , the calculation formula is: , In the formula, is the total number of occurrences of the fault type, They are The preset scaling factor of Both are greater than 0.

5. The fault prediction method for a FPSO riser support system according to claim 1, characterized in that: Obtain the historical production reduction data corresponding to each fault type, and calculate the production reduction coefficient based on the historical maintenance data, including: For each final fault type, the number of occurrences of the fault type in the historical data is obtained, and the difference between the preset production oil volume and the actual production oil volume when each fault occurs is obtained as the reduced production oil volume of that time; Add up all the reduced oil production when each failure occurs to obtain the reduction coefficient.

6. A method for predicting a fault of a FPSO riser support system according to claim 1, characterized in that: The key coefficient of each fault type is calculated based on the maintenance importance coefficient, pollution coefficient and production reduction coefficient, and the maintenance urgency of each FPSO riser support system fault type is determined based on the key coefficient, including: ； In the formula, is the critical coefficient of the fault type, They are maintenance importance coefficient, pollution coefficient and production reduction coefficient; They are The preset scaling factor of All are greater than 0; According to the size of the critical coefficient, the fault types of each FPSO riser support system are ranked from high to low in terms of maintenance urgency.

7. A fault prediction system for a FPSO riser support system, used to implement a fault prediction method for a FPSO riser support system according to any one of claims 1 to 6, characterized in that: The system comprises: Final fault type module: obtains data on the fault status of the FPSO riser support system as current fault data, calculates the similarity with each type of historical fault data of the FPSO riser support system, and predicts the fault type of the FPSO riser support system as the final fault type; Maintenance importance module: For each final fault type, obtain the historical maintenance data corresponding to each fault type, and calculate the maintenance importance coefficient based on the historical maintenance data; Pollution module: obtains historical pollution data corresponding to each fault type and calculates the pollution coefficient based on the historical pollution data; Production reduction module: obtains historical production reduction data corresponding to each fault type, and calculates the production reduction coefficient based on historical maintenance data; Maintenance module: Calculate the key coefficient of each fault type based on the maintenance importance coefficient, pollution coefficient and production reduction coefficient, and judge the maintenance urgency of the fault type of each FPSO riser support system based on the key coefficient.

Images