A method and system for evaluating the health status of offshore equipment

Abstract

The present invention relates to the technical field of equipment status assessment, and in particular to a method and system for assessing the health status of offshore operating equipment. The method comprises the following steps: acquiring offshore operating equipment component data; performing initial life marking on the offshore operating equipment component data to generate initial life marking data of the offshore operating equipment component; performing real-time operation data acquisition on the offshore operating equipment component data to generate real-time operation data of the offshore operating equipment component; performing component division on the real-time operation data of the offshore operating equipment component based on the components of the initial life marking data to generate operation data of new components and operation data of old components; performing environmental change impact analysis on the operation data of the new components to generate external change impact factors of the new components. The present invention improves the accuracy of the health status assessment of offshore operating equipment through multi-dimensional data acquisition, environmental and electrical impact analysis, fatigue damage assessment and intelligent prediction.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment status assessment, and in particular to a method and system for assessing the health status of offshore operating equipment. Background Art

[0002] Early equipment maintenance mainly relied on regular inspections and empirical judgments, lacking effective health status monitoring methods, resulting in frequent equipment failures, production line stagnation, and huge losses. With the development of industrial automation and information technology, especially the popularization of sensor technology, equipment health status assessment has gradually transformed into data-driven. In the 1990s, fault diagnosis and predictive maintenance began to emerge, using vibration, temperature, pressure and other sensors to monitor the operating status of equipment, combined with statistical methods to analyze equipment failure modes, and provide a basis for maintenance decisions. The research at this stage mainly focused on data acquisition and basic analysis technology. After entering the 21st century, with the development of big data, artificial intelligence and machine learning technologies, equipment health assessment methods have been continuously upgraded. From traditional monitoring based on a single sensor to complex assessment models with multi-source data fusion. However, the current traditional equipment health assessment methods often rely only on fixed historical data or simple sensor monitoring, ignoring the aging process of components at different stages. At the same time, the health assessment of old components often relies only on fatigue damage analysis, but fatigue damage cannot fully reveal the overall health status of the equipment, which leads to low accuracy in health status assessment of offshore equipment. Summary of the invention

[0003] Based on this, it is necessary to provide a method and system for evaluating the health status of offshore operating equipment to solve at least one of the above technical problems.

[0004] To achieve the above object, a method for evaluating the health status of offshore equipment is provided, the method comprising the following steps:

[0005] Step S1: Acquire data of offshore operating equipment components; perform initial life marking on the offshore operating equipment component data to generate initial life marking data of offshore operating equipment components; perform real-time operation data collection on the offshore operating equipment component data to generate real-time operation data of the offshore operating equipment components;

[0006] Step S2: Based on the components of the initial life mark data, the real-time operation data of the offshore equipment components are divided into components to generate the operation data of new components and the operation data of old components; the environmental change impact analysis is performed on the operation data of the new components to generate the external change impact factors of the new components; the electrical system health assessment is performed on the operation data of the new components to generate the internal change impact factors of the new components; the external change impact factors of the new components and the internal change impact factors of the new components are integrated to generate the health impact data of the new components;

[0007] Step S3: performing fatigue damage analysis on the operation data of the old type components to generate fatigue damage data of the old type components; predicting the health status of the offshore operating equipment based on the health impact data of the new type components and the fatigue damage data of the old type components to generate health status prediction data of the operating components;

[0008] Step S4: construct maintenance decisions based on the health status prediction data of the operating components to generate health status decisions for offshore operating equipment.

[0009] The present invention provides basic data support for subsequent health status assessment by obtaining the initial life mark data and real-time operation data of the components of offshore operating equipment. This step can ensure detailed tracking and recording of the operating status of each component, help establish a complete equipment health file, and thus lay a good data foundation for accurate prediction and decision-making. By dividing new components from old components, the state characteristics of different components can be accurately distinguished to avoid confusion and improve the accuracy of analysis. The impact of environmental changes and electrical system health assessments on new components help to reveal external and internal change factors in equipment operation, providing a more comprehensive perspective for health assessment. By integrating these influencing factors, the operating health status of components can be better understood and their potential risks can be predicted. By performing fatigue damage analysis on old components, the aging process and potential failure risks of components can be more accurately assessed. The combination of health impact data of new components for equipment health status prediction can not only evaluate the current status, but also predict potential problems in the future, which helps to identify equipment failures in advance and take appropriate measures, thereby reducing equipment failures and downtime. Based on the health status prediction data of operating components, scientific maintenance decisions can be generated. By optimizing maintenance timing and maintenance strategies, equipment downtime is reduced, maintenance costs are reduced, equipment life is extended, and production efficiency is improved. This step helps to achieve preventive maintenance, avoids the traditional passive method of relying on experience or regular maintenance, and ensures the optimal operating state of the equipment. Therefore, the present invention improves the accuracy of health status assessment of offshore equipment through multi-dimensional data collection, environmental and electrical impact analysis, fatigue damage assessment and intelligent prediction.

[0010] Preferably, step S1 comprises the following steps:

[0011] Step S11: Acquire offshore operation equipment component data;

[0012] Step S12: preprocessing the offshore equipment component data to generate standard offshore equipment component data, wherein the data preprocessing includes data cleaning, data denoising, data missing value filling and data standardization;

[0013] Step S13: Perform component life cycle analysis on the standard offshore equipment component data to generate component life cycle data; perform initial life marking on the standard offshore equipment component data based on the component life cycle data to generate initial life marking data for the offshore equipment component;

[0014] Step S14: using multi-source sensors to collect operating data of standard offshore operating equipment components, and updating the collected operating data in real time based on a preset time interval to generate real-time operating data of offshore operating equipment components.

[0015] The present invention can eliminate outliers and noise in the original data by cleaning, denoising, filling missing values ​​and standardizing the data of offshore equipment components, ensuring the accuracy and reliability of subsequent analysis and decision-making, thereby improving the overall data quality of the entire production system. Through the full life cycle analysis of components, the use status of the equipment at each stage can be deeply understood, so as to make accurate life prediction and extend the service life of the equipment. Based on this, potential failure risks can be identified in advance and preventive maintenance measures can be taken to avoid production interruptions and large-scale failures. The operation data of offshore equipment components are collected in real time by multi-source sensors, and the data is updated in combination with preset time intervals, so as to realize real-time monitoring of the equipment status during the production process. Through real-time data, abnormal conditions in the operation of the equipment, such as excessive temperature, abnormal vibration, etc., can be discovered in time, avoiding the accumulation and deterioration of equipment problems. The combination of the initial life mark of the component and the life cycle management data makes the management of the equipment more systematic and digital, reduces manual intervention, and enhances the level of intelligent management. Operators can accurately formulate maintenance plans based on data to avoid blind repairs and unnecessary downtime. Through accurate prediction of equipment component life and real-time operation data, personalized maintenance plans can be formulated according to the actual status of the equipment, rather than performing maintenance according to fixed time cycles, thereby effectively reducing excessive maintenance and maintenance costs and improving resource utilization efficiency.

[0016] Preferably, performing component life cycle analysis on standard offshore equipment component data includes:

[0017] Conduct demand analysis and modeling on the data of standard offshore equipment components to generate functional demand model data; conduct process design simulation on the functional demand model data to generate component design phase data;

[0018] Optimize the processing flow of the component design phase data to generate processing flow optimization data; perform quality control inspection on the processing flow optimization data to generate quality inspection report data; perform assembly accuracy analysis on the quality inspection report data to generate component production phase data;

[0019] Conduct operation monitoring and analysis on component production phase data to generate health status assessment data; conduct fault diagnosis analysis on health status assessment data to generate fault diagnosis report data; perform performance optimization and adjustment on fault diagnosis report data to generate component use phase data;

[0020] Perform periodic maintenance analysis on component use phase data to generate maintenance plan data; perform repair effect verification on maintenance plan data to generate component maintenance phase data; perform retirement assessment analysis on component maintenance phase data to generate retirement assessment data; perform disassembly process analysis on retirement assessment data to generate component retirement phase data;

[0021] Based on the data from the component design stage, component production stage, component usage stage, component maintenance stage and component retirement stage, the data of the entire life cycle of equipment operation is integrated to generate component full life cycle data.

[0022] The present invention ensures that the design stage can fully meet the actual use requirements through demand analysis and functional demand modeling of standard offshore equipment component data, and identifies potential design defects in advance through process design simulation, optimizes the design scheme, and improves the reliability and functionality of the product. Process design simulation and process optimization ensure the optimal configuration of each link in the production process, reducing waste, errors and resource waste in the production process. Quality control detection and precision analysis further ensure high precision and high pass rate in the processing process, and improve production efficiency and yield rate. Quality inspection report data and assembly precision analysis provide a comprehensive quality assessment mechanism to ensure that each production link is strictly tested. Through quality traceability, the production quality of each component can be tracked, the reliability of the product is enhanced, and rework and maintenance are reduced. Operation monitoring analysis and health status assessment can monitor the operating status of components in real time, predict the health status of components, detect abnormalities in time, and reduce the risk of downtime and damage. Fault diagnosis analysis provides a basis for fault prevention and significantly improves the continuity and stability of the production line. Through fault diagnosis reports and performance optimization adjustments, not only can the components with degraded performance be discovered and adjusted in time, but also the maintenance strategy can be optimized through periodic maintenance analysis and repair effect verification, extending the service life of the equipment and reducing maintenance costs.

[0023] Preferably, step S2 comprises the following steps:

[0024] Step S21: dividing the real-time operation data of the offshore equipment components based on the components of the initial life mark data to generate new component operation data and old component operation data;

[0025] Step S22: collecting operating environment noise of the new component operating data to obtain operating environment noise data of the new component; performing noise separation on the operating environment noise data of the new component to generate operating noise of the new component and environmental separation noise; performing environmental change impact analysis on the operating noise of the new component according to the environmental separation noise to generate external change impact factors of the new component;

[0026] Step S23: Screening the operation data of the new component motor to obtain the operation data of the new component motor; performing current and voltage frequency analysis on the operation data of the new component motor to generate the current frequency data of the new component motor and the voltage frequency data of the new component motor; performing electrical system health assessment on the current frequency data of the new component motor and the voltage frequency data of the new component motor to generate the internal change influencing factor of the new component motor;

[0027] Step S24: Integrate the data of the external change influencing factors of the new component and the internal change influencing factors of the new component to generate health impact data of the new component.

[0028] The present invention can clearly distinguish the operation conditions of new and old components by dividing the real-time operation data of offshore equipment components based on the initial life mark data, thereby realizing more accurate differentiated management. Independent monitoring of new and old components can formulate specific maintenance and optimization strategies according to the characteristics of different components. The collection and noise separation of the operating environment noise of new components can effectively analyze the impact of the environment on equipment performance. Through the analysis of environmental change influencing factors, the specific impact of external environmental factors on component performance can be quantified, providing a scientific basis for the optimal design and maintenance of equipment, and reducing the negative impact of the external environment on component life. By screening the equipment motor operation data and analyzing the current and voltage frequency, the operating status of the electrical system can be accurately obtained. The electrical system health assessment can timely discover potential problems in the motor system, such as electrical faults or abnormal fluctuations, take measures in advance to avoid equipment failures, and ensure the stable operation of the equipment. The external change influencing factors of new components are integrated with the internal change influencing factors to provide a comprehensive assessment of the health status of the components. This comprehensive assessment can not only detect the operating status of the components, but also identify potential hidden dangers of failure, making maintenance decisions more accurate and scientific. By combining external environmental influences and internal electrical system health assessments, it is possible to predict potential problems of components in actual operation in advance and develop targeted fault prevention measures. This accurate predictive capability can effectively avoid the occurrence of equipment failures, thereby reducing production interruptions, downtime and maintenance costs.

[0029] Preferably, step S22 includes the following steps:

[0030] Step S221: collecting operating environment noise of the new component operating data to obtain the operating environment noise data of the new component; performing noise signal time-frequency analysis on the operating environment noise data of the new component to generate operating environment noise signal time-frequency data;

[0031] Step S222: performing a signal peak value change trend analysis on the operating environment noise signal according to the operating environment noise signal time-frequency data to generate signal peak value change trend data, wherein the signal peak value change trend data includes peak value gentle change trend data and peak value fluctuation change trend data;

[0032] Step S223: Separate the new component operation noise from the operating environment noise signal time-frequency data by using the peak gentle change trend data to obtain the new component operation noise; separate the environmental noise from the operating environment noise signal time-frequency data by using the peak fluctuation change trend data to obtain the new component operation noise;

[0033] Step S224: analyzing the impact of component operation noise changes on the new component operation noise according to the environmental separation noise, and generating external change impact factors of the new component.

[0034] The present invention can accurately separate the operation noise and environmental noise of the new component through time-frequency analysis of the operation environment noise data. This process eliminates the interference of environmental noise, makes the operation noise data of the new component purer, provides a higher quality signal for subsequent analysis, and ensures the accuracy and reliability of the monitoring data. Through the signal peak change trend analysis, the change trend of the environmental noise signal can be quantitatively analyzed, and the different characteristics of the peak gentle change and the fluctuating change can be identified, so as to better understand the influence of environmental factors on the equipment noise, which helps to accurately identify and distinguish the influence of external environmental changes on the equipment operation noise, and provide a basis for the optimization design and operation and maintenance of the equipment. Through the accurate analysis of the peak change trend and noise separation, the operation noise of the new component can be more independent of the environmental noise, thereby improving the tolerance of the system to noise interference, which helps to maintain the efficiency and accuracy of equipment monitoring in noisy or complex production environments. The change impact analysis of the operation noise after the environmental noise separation helps to identify the influencing factors of environmental changes on equipment performance, such as temperature fluctuations, humidity changes, vibrations and other external factors. This analysis can provide a reference for the design of new components and the optimization of operating conditions, improve the environmental adaptability of the equipment, and reduce the negative impact of the external environment on equipment performance.

[0035] Preferably, step S23 includes the following steps:

[0036] Step S231: filtering the equipment motor operation data of the new component operation data to obtain the equipment motor operation data;

[0037] Step S232: extracting current parameters and voltage parameters from the equipment motor operation data to obtain equipment motor operation current parameters and equipment motor operation voltage parameters; performing fast Fourier transform on the equipment motor operation current parameters and equipment motor operation voltage parameters to generate equipment motor operation current spectrum diagram and equipment motor operation voltage spectrum diagram;

[0038] Step S233: performing spectrum analysis on the equipment motor running current spectrum diagram and the equipment motor running voltage spectrum diagram to generate equipment motor current frequency data and equipment motor voltage frequency data; overlapping the equipment motor current frequency data and the equipment motor voltage frequency data to generate an equipment motor running spectrum overlapping curve;

[0039] Step S234: Calculate the spectrum fluctuation mean of the equipment motor operation spectrum overlap curve to obtain the equipment motor operation spectrum mean data; screen the spectrum fluctuation influencing parameters of the equipment motor operation data based on the equipment motor operation spectrum mean data to generate the internal change influencing factors of the new component.

[0040] The present invention can extract key data closely related to motor performance by screening the equipment motor operation data, and provide accurate basic data for subsequent spectrum analysis and health assessment. This step effectively removes interference factors and ensures high quality and high precision of subsequent analysis. By extracting the motor operation current and voltage parameters, the electrical characteristics of the motor can be deeply analyzed. These parameters reflect the load condition, operation stability and health status of the electrical system of the motor, and provide an important basis for diagnosing motor performance problems. By performing fast Fourier transform (FFT) on the equipment motor current and voltage data, a spectrum diagram of motor operation can be generated. Spectral analysis can reveal the frequency characteristics during the operation of the motor, help identify abnormalities such as vibration, noise, harmonics, etc. during the operation of the equipment, and discover potential faults and performance degradation in advance. Overlapping the spectrum curves of the motor current frequency data and the motor voltage frequency data can fully understand the electrical system status of the motor. By integrating different data sources, more comprehensive equipment status information can be provided, thereby improving the diagnostic capability of the equipment health status. By calculating the fluctuation mean of the equipment motor operation spectrum, the stability of the motor operation and its health status can be effectively evaluated. The mean value of spectrum fluctuations can reflect periodic changes or abnormal fluctuations in equipment operation, providing an important reference for equipment performance degradation or fault diagnosis. The spectrum fluctuation influencing parameters screened out based on spectrum mean data can provide a strong basis for continuous monitoring of motor operation status. Timely detection of abnormal changes in spectrum fluctuations can provide early warning of potential equipment failures, thereby reducing downtime and maintenance costs.

[0041] Preferably, step S3 comprises the following steps:

[0042] Step S31: performing fatigue damage analysis on the old component operation data to generate old component fatigue damage data;

[0043] Step S32: allocating state weights to the health impact data of new components and fatigue damage data of old components based on the entropy weight method to generate state weight data of components of offshore equipment;

[0044] Step S33: dividing the offshore equipment component status weight data into a data set to generate a model training set and a model test set; using a convolutional neural network algorithm to perform model training on the model training set to generate a pre-model for predicting the health status of offshore equipment components;

[0045] Step S34: perform model optimization iteration on the pre-model for predicting the health status of offshore operating equipment components through the model test set, thereby generating a health status prediction model for operating components; import the offshore operating equipment component status weight data into the operating component health status prediction model to predict the health status of offshore operating equipment, and generate operating component health status prediction data.

[0046] The present invention can effectively evaluate the service life and damage degree of old components and identify fatigue points by performing fatigue damage analysis on the operation data of old components, which provides data support for component maintenance decisions and helps avoid failures caused by component fatigue damage, thereby improving the operational reliability of the equipment. Combining the health impact data of new components with the fatigue damage data of old components, the state weight distribution is performed through the entropy weight method, and the health status of each component of offshore equipment can be comprehensively evaluated. This comprehensive evaluation helps to accurately identify which components need priority attention, help formulate reasonable maintenance and replacement strategies, and avoid overall equipment failure due to poor health of some components. The health status of offshore equipment components is predicted by the convolutional neural network (CNN) algorithm, which can automatically learn and extract key features from a large amount of historical data, thereby accurately predicting the health status of the equipment. Using deep learning technology, the model can be continuously optimized and iterated to improve the accuracy and stability of the prediction. By constructing a health status prediction model for offshore equipment components, the future health status of the equipment can be predicted, and potential failures and performance degradation risks can be identified in advance. This provides a powerful fault warning mechanism for enterprises, which can take preventive measures before serious equipment failures occur and reduce the risk of sudden downtime. By dividing and optimizing the training set and test set, the health status of the equipment can be accurately classified and predicted. This process reduces the interference of human factors and improves the model's ability to identify the status of different components, thereby ensuring comprehensive control of the health status of offshore equipment components.

[0047] Preferably, step S31 includes the following steps:

[0048] Step S311: performing stress distribution calculation on the old component operation data to generate the old component operation stress distribution data; performing cyclic loading extraction on the old component operation stress distribution data to obtain a cyclic loading data set; performing load frequency statistics on the cyclic loading data set to generate load frequency distribution data;

[0049] Step S312: analyzing the material characteristics of the old-type component on the old-type component operation data to generate the material characteristic data of the old-type component; constructing the SN curve on the material characteristic data of the old-type component to generate the SN curve of the old-type component;

[0050] Step S313: Based on Miner's linear damage accumulation theory, regional damage assessment is performed on the load frequency distribution data and the SN curve of the old component to generate fatigue damage data of the old component.

[0051] The present invention can understand the stress condition of the component in actual operation in detail by calculating the stress distribution of the operation data of the old component, identify potential high stress areas, and provide basic data for subsequent fatigue analysis and fault prediction. This step provides a quantitative basis for the structural design and performance optimization of the component, which helps to find design defects and potential faults in advance. By extracting the stress distribution data by cyclic loading, the key data in the load cycle can be extracted, and accurate information is provided for load frequency statistics and fatigue analysis. This process helps to refine the stress conditions of the component under different working conditions, which helps to more accurately evaluate the fatigue life of the component. Through the generation of load frequency distribution data, the load changes borne by the component during the actual operation can be analyzed, which helps to reveal the stress mode of the component under different load conditions, further optimize the design and improve the service life of the component, and reduce the damage caused by excessive load. Analyzing the material properties of the old component can deeply understand its bearing capacity under different stress conditions. Based on the construction of the SN curve, the fatigue life of the component under cyclic load can be accurately predicted, providing reliable theoretical support for component design and optimization, and helping to avoid structural damage caused by insufficient material properties. By applying Miner's linear damage accumulation theory, the fatigue damage accumulation process of components under different load frequencies can be quantitatively evaluated. This method can accurately predict the fatigue life of components, help formulate reasonable maintenance strategies, and reduce sudden failures caused by fatigue damage. By calculating the fatigue damage data of old components, the fatigue life of components can be predicted in advance, and maintenance or replacement can be performed before the components reach the critical fatigue value. This can effectively prevent equipment failures caused by fatigue cracking and reduce the risk of production interruptions and equipment downtime.

[0052] Preferably, step S4 comprises the following steps:

[0053] Step S41: Visualizing the health status of the operating components based on their health status prediction data to generate a health status display interface for offshore operating equipment;

[0054] Step S42: constructing maintenance decisions for the health status prediction data of the operating components according to the health status display interface of the offshore operating equipment, and generating a health status decision for the offshore operating equipment.

[0055] The present invention visualizes the health status of production components, so that users can intuitively view the real-time health status of the equipment. This visualization not only improves the operator's understanding of the equipment status, but also provides decision support for maintenance personnel, helping them to make corresponding maintenance decisions in time and avoid the occurrence of potential failures. Maintenance decisions based on health status prediction data are helpful to optimize maintenance processes and resource allocation. By predicting the health status of equipment in advance, enterprises can plan maintenance work in a planned manner, thereby reducing emergency repairs and downtime and improving production efficiency. Through real-time display and predictive analysis of health status, potential equipment problems can be identified in advance, reducing the risk of production stagnation or equipment damage caused by equipment failure. Timely maintenance measures can effectively avoid the occurrence of serious failures and extend the service life of equipment. Through the combination of health status prediction data and maintenance decisions, production plans can be adjusted more accurately. Enterprises can arrange production tasks according to the health status of equipment, avoid arranging excessive loads during high-risk periods of equipment, and ensure the smooth progress of the production process. Health status visualization not only provides intuitive feedback on equipment health, but also provides data-driven decision-making basis for management through decision support systems, which can optimize equipment management processes, improve decision quality, and reduce human errors.

[0056] In this specification, a system for evaluating the health status of offshore equipment is provided, which is used to execute the above-mentioned method for evaluating the health status of offshore equipment. The system for evaluating the health status of offshore equipment includes:

[0057] The data acquisition module is used to obtain the data of the components of the offshore operating equipment; to perform initial life marking on the data of the components of the offshore operating equipment to generate the initial life marking data of the components of the offshore operating equipment; to perform real-time operation data collection on the data of the components of the offshore operating equipment to generate the real-time operation data of the components of the offshore operating equipment;

[0058] The new equipment analysis module is used to divide the real-time operation data of offshore equipment components based on the components with initial life mark data, and generate the operation data of new components and old components; perform environmental change impact analysis on the operation data of new components, and generate the external change impact factors of new components; perform electrical system health assessment on the operation data of new components, and generate the internal change impact factors of new components; integrate the external change impact factors of new components and the internal change impact factors of new components to generate the health impact data of new components;

[0059] The old equipment analysis module is used to perform fatigue damage analysis on the operation data of old components and generate fatigue damage data of old components; predict the health status of offshore operating equipment based on the health impact data of new components and fatigue damage data of old components and generate health status prediction data of operating components;

[0060] The health decision module is used to build maintenance decisions based on the health status prediction data of operating components and generate health status decisions for offshore operating equipment.

[0061] The beneficial effect of the present invention is that by collecting the operating data of the components of offshore equipment in real time, it is possible to ensure dynamic monitoring of the equipment status, timely discover potential faults or performance degradation, and reduce the risk of unexpected equipment downtime. By marking the component data with initial life and dividing it into new components and old components, it is helpful to adopt differentiated maintenance strategies for components with different life cycles, which not only improves the accuracy of maintenance, but also enables targeted management based on the actual conditions of the components. Environmental change impact analysis and electrical system health assessment of new components can identify the impact of external environment and internal changes of components on equipment health. This multi-dimensional analysis method helps to comprehensively evaluate the health status of the equipment and then formulate more accurate prediction and maintenance strategies. Fatigue damage analysis of old components can identify potential damage accumulation areas in advance and help predict the remaining service life of components. Predicting the health status of offshore equipment in combination with the health impact data of new components can improve the accuracy of fault warnings and reduce downtime and maintenance costs caused by equipment failures. Maintenance decision-making based on equipment health status prediction data can optimize maintenance plans and ensure that repairs or parts are replaced when the equipment is in optimal working condition, thereby reducing unnecessary downtime and excessive maintenance and improving overall production efficiency and equipment utilization. Through accurate health status prediction and maintenance decisions, the service life of equipment can be extended, the risk of unexpected downtime in the production process can be reduced, and it also helps to control maintenance costs and improve resource utilization efficiency. Therefore, the present invention improves the accuracy of health status assessment of offshore equipment through multi-dimensional data collection, environmental and electrical impact analysis, fatigue damage assessment and intelligent prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 A schematic diagram of the steps of a method for evaluating the health status of offshore equipment;

[0063] Figure 2 for Figure 1 Detailed implementation steps of step S2 in the flowchart;

[0064] Figure 3 for Figure 1 Detailed implementation steps of step S3 in FIG.

[0065] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0066] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are 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 technicians in this field without creative work are within the scope of protection of the present invention.

[0067] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0068] It should be understood that although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit, and the term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0069] To achieve this, please refer to Figures 1 to 3 , a method for evaluating the health status of offshore equipment, the method comprising the following steps:

[0070] Step S1: Acquire data of offshore operating equipment components; perform initial life marking on the offshore operating equipment component data to generate initial life marking data of offshore operating equipment components; perform real-time operation data collection on the offshore operating equipment component data to generate real-time operation data of the offshore operating equipment components;

[0071] Step S2: Based on the components of the initial life mark data, the real-time operation data of the offshore equipment components are divided into components to generate the operation data of new components and the operation data of old components; the environmental change impact analysis is performed on the operation data of the new components to generate the external change impact factors of the new components; the electrical system health assessment is performed on the operation data of the new components to generate the internal change impact factors of the new components; the external change impact factors of the new components and the internal change impact factors of the new components are integrated to generate the health impact data of the new components;

[0072] Step S3: performing fatigue damage analysis on the operation data of the old type components to generate fatigue damage data of the old type components; predicting the health status of the offshore operating equipment based on the health impact data of the new type components and the fatigue damage data of the old type components to generate health status prediction data of the operating components;

[0073] Step S4: construct maintenance decisions based on the health status prediction data of the operating components to generate health status decisions for offshore operating equipment.

[0074] The present invention provides basic data support for subsequent health status assessment by obtaining the initial life mark data and real-time operation data of the components of offshore operating equipment. This step can ensure detailed tracking and recording of the operating status of each component, help establish a complete equipment health file, and thus lay a good data foundation for accurate prediction and decision-making. By dividing new components from old components, the state characteristics of different components can be accurately distinguished to avoid confusion and improve the accuracy of analysis. The impact of environmental changes and electrical system health assessments on new components help to reveal external and internal change factors in equipment operation, providing a more comprehensive perspective for health assessment. By integrating these influencing factors, the operating health status of components can be better understood and their potential risks can be predicted. By performing fatigue damage analysis on old components, the aging process and potential failure risks of components can be more accurately assessed. The combination of health impact data of new components for equipment health status prediction can not only evaluate the current status, but also predict potential problems in the future, which helps to identify equipment failures in advance and take appropriate measures, thereby reducing equipment failures and downtime. Based on the health status prediction data of operating components, scientific maintenance decisions can be generated. By optimizing maintenance timing and maintenance strategies, equipment downtime is reduced, maintenance costs are reduced, equipment life is extended, and production efficiency is improved. This step helps to achieve preventive maintenance, avoids the traditional passive method of relying on experience or regular maintenance, and ensures the optimal operating state of the equipment. Therefore, the present invention improves the accuracy of health status assessment of offshore equipment through multi-dimensional data collection, environmental and electrical impact analysis, fatigue damage assessment and intelligent prediction.

[0075] In the embodiment of the present invention, reference Figure 1 FIG. 1 is a schematic diagram of a step flow chart of a method for evaluating the health status of offshore equipment according to the present invention. In this example, the method for evaluating the health status of offshore equipment includes the following steps:

[0076] Step S1: Acquire data of offshore operating equipment components; perform initial life marking on the offshore operating equipment component data to generate initial life marking data of offshore operating equipment components; perform real-time operation data collection on the offshore operating equipment component data to generate real-time operation data of the offshore operating equipment components;

[0077] In an embodiment of the present invention, by installing sensors (such as vibration sensors, temperature sensors, pressure sensors, etc.) on the equipment, the operation data of the equipment components are collected. A data acquisition system (such as a PLC system, a SCADA system, etc.) is used for centralized data acquisition to ensure the real-time and accuracy of the data. According to the acquisition frequency and transmission mode of the sensor, a suitable data acquisition technology (such as wireless transmission, real-time network transmission, etc.) is selected for data synchronization. Once per second or appropriately adjusted according to the operating rate of the equipment. For high-speed equipment, the acquisition frequency is set to 100 times per second to ensure the accuracy of the data. Different types of sensors are selected according to the parameters to be monitored. For example, a temperature sensor (such as a thermocouple) is used to detect the temperature change of the equipment components, and a vibration sensor (such as an accelerometer) is used to detect the vibration state of the equipment. Based on the technical manual or usage standards provided by the manufacturer of the equipment, combined with the historical data of the equipment, the initial state of each equipment component is determined. The components are tested for the first operation, and the initial operating parameters (such as temperature, speed, load, etc.) are recorded and used as initial life data. Combined with the equipment's use environment, installation conditions, etc., determine the equipment's initial life cycle and assign an initial life mark. The initial life mark of a motor can be set to "1000 hours", which is based on a comprehensive assessment of the manufacturer's test results and environmental factors. The initial life parameters include the initial health status of the component, operating hours, temperature, vibration and other data. Specifically, the initial life mark of the motor is "0 hours" and is recorded as "10 hours" after the first operation. During the operation of the equipment, the sensor will continuously monitor and record various operating parameters of the equipment, such as temperature, vibration, load, etc. Real-time data is transmitted to the server for analysis through the data acquisition system. A data acquisition and monitoring system (such as an IoT platform or an industrial big data platform) is used to achieve real-time transmission, storage and processing of data. The data acquisition system should have high real-time performance and be able to respond to changes in the equipment's operating status in a timely manner. For the acquisition of real-time data, the acquisition frequency is usually once per second to ensure that short-term fluctuations in the operation of the equipment are captured. For high-precision equipment, the acquisition frequency can be increased to 50 times per second. Operating parameters include temperature, vibration, pressure, power, speed, etc., which monitor the working status of the equipment in real time. The real-time temperature of the motor can be collected through the temperature sensor, and the vibration data can be obtained through the accelerometer.

[0078] Step S2: Based on the components of the initial life mark data, the real-time operation data of the offshore equipment components are divided into components to generate the operation data of new components and the operation data of old components; the environmental change impact analysis is performed on the operation data of the new components to generate the external change impact factors of the new components; the electrical system health assessment is performed on the operation data of the new components to generate the internal change impact factors of the new components; the external change impact factors of the new components and the internal change impact factors of the new components are integrated to generate the health impact data of the new components;

[0079] In the embodiment of the present invention, the new components and old components in the equipment are distinguished by using the initial life mark data. The basis for this division is usually the relationship between the time the component is put into use and its expected design life. The specific division method is as follows: If the cumulative use time of the component in the equipment is less than half of the initial life mark time of the component, it is classified as a new component. If the cumulative use time of the component in the equipment is greater than or equal to half of the initial life mark time, it is classified as an old component. The initial life mark time (for example, 1000 hours) of each component is obtained by querying the database and analyzing the operation history of the equipment. The relationship between the current use time of the equipment and the initial life mark time is compared, the data of the new components and the old components are divided, and the corresponding generated operation data is classified to ensure that the data of the new components and the old components are correctly allocated. Environmental sensors are deployed at the equipment operation site to monitor environmental variables such as temperature, humidity, and air pressure in real time. Assuming that the equipment is in an outdoor environment, these environmental factors directly affect the health of the equipment components. Temperature sensor: measures the temperature of the working environment (for example: 30°C). Humidity sensor: measures the ambient humidity (for example: 50%). Pressure sensor: measures the air pressure (for example: 1013hPa). Based on the relationship between environmental data and equipment operation data, regression analysis, machine learning models (such as linear regression, support vector machine, etc.) or empirical formulas are used to analyze how environmental factors affect the operating status of components. Every 1°C change in temperature leads to a 0.5% reduction in equipment life. The comprehensive impact factor of temperature, humidity and air pressure can be calculated by weighted average or linear model, assuming that the impact of temperature, humidity and air pressure on the equipment is -0.3%, -0.2% and -0.1% respectively. Environmental impact factor = (temperature impact factor × w1) + (humidity impact factor × w2) + (air pressure impact factor × w3) where w1, w2, w3 are the weight coefficients of temperature, humidity and air pressure (for example: 0.5, 0.3, 0.2). The electrical operation data of new components is collected in real time through current and voltage sensors. Monitor the voltage of the component (for example: 220V), monitor the current of the component (for example: 10A), and calculate the power (for example: 2.2kW) based on the voltage and current data. Use a condition monitoring model (such as a degradation model based on the electrical system) to evaluate the health of the electrical system of the component. If the current exceeds the normal range (for example, the fluctuation exceeds 10%), the electrical system is considered abnormal. According to the health status of the electrical system, use empirical formulas or machine learning models to calculate the internal variation factor. For example, when the current fluctuation exceeds 10%, the health impact factor of the electrical system is -0.4%. Generate the internal variation impact factor of the new component based on the results of the electrical system health assessment and store it in the data management system. Use the weighted average method or other methods to combine the external variation factor with the internal variation factor. For example, environmental factors have a greater impact on the equipment, and a higher weight is given to the external factor. Assume that the weights of the external and internal factors are 0.6 and 0.4, respectively.The health impact data can be calculated using the following formula: health impact data = (external impact factor × 0.6) + (internal impact factor × 0.4); the calculated health impact data is stored in the database for subsequent analysis and maintenance decision-making.

[0080] Step S3: performing fatigue damage analysis on the operation data of the old type components to generate fatigue damage data of the old type components; predicting the health status of the offshore operating equipment based on the health impact data of the new type components and the fatigue damage data of the old type components to generate health status prediction data of the operating components;

[0081] In an embodiment of the present invention, the operating data of the old components are collected in detail, especially the physical quantities related to fatigue damage: the maximum stress value during operation is collected by a stress sensor (for example, 100 MPa). The deformation data of the component is collected by a strain sensor (for example, the strain value is 0.002). The load size and fluctuation are collected (for example, the load varies between 500N and 1500N). Common fatigue damage models, such as Miners' law or SN curve model, are used to estimate the degree of damage of the component. Miners' law is used to calculate the total fatigue damage under multiple cyclic loads, and the formula is: ;in is the total damage degree, For the Number of cycles under load, For the Fatigue life under load (i.e. the maximum number of cycles a component can withstand under this load condition). Use the material's SN curve (stress-life curve) to estimate the fatigue life of a component by measuring the stress during operation: ;in is the fatigue life, and is the material constant, is the stress amplitude. Based on the above fatigue damage model, the fatigue damage degree of the old components is calculated to generate fatigue damage data. Assume that the calculated damage degree is 30%, that is, the component has been damaged by 30%. Select a suitable health status prediction model, such as the weighted average method, machine learning model (such as support vector machine SVM, random forest, neural network, etc.) or multi-factor prediction model. Here, the weighted average method is used as an example for detailed explanation. Combine the health impact data of the new components with the fatigue damage data of the old components. We set the weighting coefficient to reflect the relative importance of the new and old components to the health status of the equipment. For example, the weight of the health impact factor of the new component is 0.4, and the weight of the fatigue damage factor of the old component is 0.6. Weighted calculation formula: Equipment health status = (new component health impact factor × 0.4) + (old component fatigue damage factor × 0.6); the new component health impact factor is -0.3%, the old component fatigue damage factor is 30%, then the predicted value of the equipment health status is: Equipment health status = (-0.3% × 0.4) + (30% × 0.6) = -0.12% + 18% = 17.88%; classify and evaluate the health status according to the prediction results, and set different health status thresholds: Healthy: The equipment health status is less than 10% damage. Medium: The equipment health status is between 10% and 30% damage. Unhealthy: The equipment health status is greater than 30% damage. In this example, the equipment health status is 17.88%, which is a "medium" state and requires regular inspections or small-scale maintenance.

[0082] Step S4: construct maintenance decisions based on the health status prediction data of the operating components to generate health status decisions for offshore operating equipment.

[0083] In the embodiment of the present invention, the health status prediction data of the equipment is used as input, and these data reflect the health status of each component, such as the damage degree, fatigue degree and health influencing factor of the component. Different health status thresholds are set according to the evaluation results of the health status of the equipment. Generally, the health status prediction data of the equipment are divided into several levels, and each level corresponds to a different maintenance response. For example: Health status ≤ 10%: the equipment is in good health and does not require immediate maintenance. 10% < health status ≤ 30%: the equipment is in a medium health state and requires regular inspection and monitoring to avoid potential failures. Health status > 30%: the equipment is in poor health and requires emergency maintenance or repair to prevent failures. According to different health status, a corresponding maintenance decision model is constructed: Good health status (≤ 10% damage): the equipment can continue to operate, but regular monitoring and inspection need to be set, such as a health assessment once a month or every quarter. Medium health status (10%~30% damage): a detailed inspection is required within a certain period of time (such as 2 weeks or 1 month). It can be set to light repair or replacement of parts to ensure that the equipment does not have serious failures. Poor health status (>30% damage): Immediately shut down and perform emergency repairs or replace damaged parts. At this time, a comprehensive inspection and repair can be carried out to prevent significant losses caused by production line stagnation or equipment failure. Use multi-level decision trees, weighted decision methods, or rule-based decision systems to build maintenance decision models. These methods automatically make maintenance decisions according to the rules based on the input health status prediction data. Multi-level decision tree model: Input: equipment health status prediction data. Output: maintenance decisions, such as "immediate repair", "regular inspection", "continue operation", etc. Input equipment health status data and determine the degree of damage to the equipment based on the prediction results. Determine the health status threshold and divide the equipment into different health status categories. Determine the maintenance strategy based on the category, such as regular inspection, emergency maintenance, shutdown maintenance, etc. Weighted calculation of equipment health status data with different maintenance costs, production risks, resource utilization and other factors, comprehensively consider the equipment health status and the resources required for maintenance, and generate the optimal maintenance decision. According to the maintenance strategy output by the above model, start to formulate a specific maintenance plan. Arrange relevant personnel to conduct monthly or quarterly inspections. According to the evaluation results of the medium health status, arrange maintenance personnel to replace parts, and immediately repair or replace damaged parts of equipment in poor health status. Regularly track the execution of maintenance decisions and optimize the decision model through feedback mechanisms. For example, if the maintenance decision is found to be unsatisfactory in actual execution, the parameters of the decision model (such as the weight of the health status) can be adjusted or the maintenance strategy can be changed.

[0084] Preferably, step S1 comprises the following steps:

[0085] Step S11: Acquire offshore operation equipment component data;

[0086] Step S12: preprocessing the offshore equipment component data to generate standard offshore equipment component data, wherein the data preprocessing includes data cleaning, data denoising, data missing value filling and data standardization;

[0087] Step S13: Perform component life cycle analysis on the standard offshore equipment component data to generate component life cycle data; perform initial life marking on the standard offshore equipment component data based on the component life cycle data to generate initial life marking data for the offshore equipment component;

[0088] Step S14: using multi-source sensors to collect operating data of standard offshore operating equipment components, and updating the collected operating data in real time based on a preset time interval to generate real-time operating data of offshore operating equipment components.

[0089] In an embodiment of the present invention, data is collected through the intelligent sensors, PLC systems, SCADA systems, etc. of the equipment. Common data include parameters such as temperature, vibration, pressure, rotation speed, current, voltage, etc. The sensor data of each component should cover key indicators during operation, such as temperature (°C), vibration acceleration (m / s²), current (A), pressure (Pa), etc. Set a suitable acquisition frequency, such as once per second (1Hz) or dynamically adjust the acquisition frequency according to the operating characteristics of the component. For high-speed rotating components (such as motors), higher frequency data acquisition (for example, 10Hz) is required. Multi-source sensors are configured for each component of the offshore operating equipment to ensure that all types of data can be accurately collected. The sensor should be connected to the communication system of the equipment (such as CAN bus or Modbus protocol) to ensure that the data is transmitted to the data processing center in real time. Use statistical methods such as mean median method or outlier detection based on machine learning models (such as isolation forest, K-means clustering) to identify noise data. Apply low-pass filters or wavelet transforms to smooth the data and reduce the impact of high-frequency noise. Use interpolation methods (such as linear interpolation, Lagrange interpolation) or machine learning algorithms (such as K-nearest neighbor filling, regression model) to estimate missing values. Apply z-score standardization or minimum-maximum standardization to scale the data to a standard range, such as [0, 1] or [-1, 1]. Collect data on the entire process of offshore equipment components from design, manufacturing, use to scrapping to understand the influencing factors and health change trends at each stage. Life cycle data should include: component working time, cumulative operating hours, failure history, maintenance records, operating environment data (such as temperature, humidity, etc.). Use time series analysis, state space modeling and other methods to model the entire life cycle of each component. Use regression analysis and machine learning algorithms (such as random forests, neural networks) to evaluate the performance changes of components at different stages of use. Based on life cycle data, calculate the health index (HI) of each component, which reflects the health status of the component and provides a basis for subsequent life marking. Based on the historical data of the components and their operating environment, the service life prediction model (such as Weibull distribution, Exponential distribution) marks the components with initial life marks and generates initial life mark data for offshore equipment components. Use a statistically based model to determine the "initial life" of each component, that is, from the beginning of production and installation, after a certain period of operation, the component begins to enter the decline period. Configure different types of sensors (such as temperature sensors, vibration sensors, pressure sensors, etc.) to collect data in real time according to the characteristics of offshore equipment. The sampling rate of each sensor should be set according to the working characteristics and monitoring requirements of the equipment. Usually, data is collected once per second (1Hz). The sampling frequency can be increased for high-speed equipment or key components. Set the data update interval, such as updating the operating data every 10 seconds or 1 minute to ensure the timeliness of the data.Use edge computing platforms to perform preliminary processing on real-time data, filter noise, and perform preliminary diagnosis to reduce transmission load and latency. Use cloud platforms or local databases for data storage to ensure that real-time data can be called and analyzed at any time. Configure appropriate data transmission protocols (such as MQTT, HTTP / HTTPS) for reliable data transmission to ensure the integrity and security of data transmission.

[0090] Preferably, performing component life cycle analysis on standard offshore equipment component data includes:

[0091] Conduct demand analysis and modeling on the data of standard offshore equipment components to generate functional demand model data; conduct process design simulation on the functional demand model data to generate component design phase data;

[0092] Optimize the processing flow of the component design phase data to generate processing flow optimization data; perform quality control inspection on the processing flow optimization data to generate quality inspection report data; perform assembly accuracy analysis on the quality inspection report data to generate component production phase data;

[0093] Conduct operation monitoring and analysis on component production phase data to generate health status assessment data; conduct fault diagnosis analysis on health status assessment data to generate fault diagnosis report data; perform performance optimization and adjustment on fault diagnosis report data to generate component use phase data;

[0094] Perform periodic maintenance analysis on component use phase data to generate maintenance plan data; perform repair effect verification on maintenance plan data to generate component maintenance phase data; perform retirement assessment analysis on component maintenance phase data to generate retirement assessment data; perform disassembly process analysis on retirement assessment data to generate component retirement phase data;

[0095] Based on the data from the component design stage, component production stage, component usage stage, component maintenance stage and component retirement stage, the data of the entire life cycle of equipment operation is integrated to generate component full life cycle data.

[0096] In the embodiment of the present invention, through demand analysis modeling, the functional requirements of the components during the life cycle are clarified to ensure that the design and subsequent processes can meet these requirements. The functions of the equipment components are gradually disassembled and analyzed. The functional requirements of engine components include: high temperature resistance (120°C), corrosion resistance (sulfate concentration not exceeding 10%), maximum load capacity (20,000 N), and vibration frequency stability (<1Hz). The functional requirement prediction model is established using historical operation data and sensor data, and machine learning methods such as support vector machines (SVM) are used to predict the load and functional requirements of components under different working conditions. Workload data: Maximum load: 20,000N (for example, the pressure sensor measurement data of the engine) Temperature range: -10°C to 120°C (for example, component temperature resistance requirements) Ambient humidity: 25% to 85% (the impact of environmental factors on component functional requirements). Component design specifications: Strength standard: material yield strength ≥250 MPa Anti-corrosion standard: in accordance with ISO12944 anti-corrosion standard. Use the historical operation data of the equipment to verify whether the functional requirements predicted by the model are accurate. For example, the effectiveness of the model can be verified by comparing the actual operating data with the predicted values ​​under different load and environmental conditions. Through process design simulation, it is ensured that the component design can meet the functional requirements during the processing and can smoothly enter the subsequent production. Use CAD software (such as SolidWorks) to design the components in detail, simulate the processing process, and optimize the processing path. Use FEA software (such as ANSYS) to simulate the components in terms of stress, heat, vibration, etc. to verify the reliability of the design. Use process simulation platforms (such as DELMIA) to simulate the processing of components and determine the optimal processing path and process parameters. Use genetic algorithms (GA) or particle swarm optimization (PSO) to optimize processing paths, tool selection and cutting parameters. Use simulation software (such as Simul8) to simulate different process flows and select the optimal solution. The processing time before optimization is 30 minutes, and the processing time after optimization is 20 minutes. The utilization rate of the original material is 80%, and the utilization rate after optimization is 90%. The error before optimization is 0.1mm, and the error after optimization is 0.05mm. Use CMM to test the dimensional accuracy of each processed component to ensure that the component meets the design requirements. Scan the surface quality of the parts to measure surface roughness and defects. The tolerance of the outer dimensions of the parts is ±0.1mm, and the tolerance of the aperture is ±0.05mm. The Ra value is required to be ≤1.0µm, in accordance with ISO 4287. The clearance is required to be ≤0.02mm to ensure that there is no deviation in the operation after assembly. Compare the dimensional data measured by the CMM with the design requirements to verify the processing accuracy. Use a laser scanner to scan the surface and compare it with the standard value to confirm whether the surface quality meets the requirements. Use CMM to detect the fit accuracy of each component during the assembly process. Perform assembly simulation in a virtual environment to evaluate the errors that occur during the assembly process.The tolerance of the component gap is required to be ≤0.01mm to ensure the matching accuracy. The matching error of each component is measured by CMM, and the error is controlled within ≤0.02mm. Use sensors (such as temperature sensors, pressure sensors, and vibration sensors) to monitor the working status of the components in real time. Use machine learning models (such as random forests) to analyze the collected data and predict the health status of the components. The vibration frequency of the components during operation is required to be ≤2 Hz. Excessive vibration frequency means mechanical failure. The temperature of the component should not exceed 150°C. Overheating is a precursor to failure. The working pressure should be kept within 200 MPa. Excessive pressure will cause component damage.

[0097] As an example of the present invention, refer to Figure 2 As shown, in this example, step S2 includes:

[0098] Step S21: dividing the real-time operation data of the offshore equipment components based on the components of the initial life mark data to generate new component operation data and old component operation data;

[0099] Step S22: collecting operating environment noise of the new component operating data to obtain operating environment noise data of the new component; performing noise separation on the operating environment noise data of the new component to generate operating noise of the new component and environmental separation noise; performing environmental change impact analysis on the operating noise of the new component according to the environmental separation noise to generate external change impact factors of the new component;

[0100] Step S23: Screening the operation data of the new component motor to obtain the operation data of the new component motor; performing current and voltage frequency analysis on the operation data of the new component motor to generate the current frequency data of the new component motor and the voltage frequency data of the new component motor; performing electrical system health assessment on the current frequency data of the new component motor and the voltage frequency data of the new component motor to generate the internal change influencing factor of the new component motor;

[0101] Step S24: Integrate the data of the external change influencing factors of the new component and the internal change influencing factors of the new component to generate health impact data of the new component.

[0102] In the embodiment of the present invention, the real-time operation data of the equipment components are classified according to the initial life mark data of the equipment components to distinguish between new components and old components. The performance and aging conditions of different types of components are different, so they need to be analyzed separately. The operation data collected in real time by the equipment sensors (temperature, pressure, vibration, rotation speed, etc.) are combined with the initial life mark data of the equipment, and the data is cut and classified using a time series data segmentation method (such as a sliding window method). Sampling frequency: 1Hz (sensor data acquisition frequency), time window: 500 seconds (data sliding window size). According to the initial life mark data of the equipment (such as purchase date, usage time, operating hours, etc.), the data is divided into two categories: new components and old components. The new component label uses a time interval (for example, usage time ≤ 500 hours) as the division standard, and the old component label uses a time interval (for example, usage time > 500 hours) as the division standard. Compare the data of the new component and the old component to verify the rationality of the division result. Several known new and old components can be selected in the experiment for calibration to ensure accurate division. During the operation of the equipment, external environmental noise affects the operating state of the components. By collecting environmental noise and separating it from the noise of the component itself, the impact of external changes on component health can be analyzed more accurately. Use environmental noise sensors (such as microphones, accelerometers, etc.) to collect noise data in real time to obtain the environmental noise of new components under different working conditions. Noise sampling frequency: 1 kHz (collecting 1000 data points per second), noise collection time: 30 minutes (length of each collection time). Use blind source separation algorithms (BSS), such as independent component analysis (ICA) technology, to separate environmental noise and component noise. Use the ICA model to extract environmental noise (such as the working environment noise around the equipment) and noise generated by the component (such as mechanical vibration noise). Data input: 5 sensors (temperature, vibration, noise, pressure, acceleration), output data: independent components of environmental noise and component noise. Perform correlation analysis on environmental noise data and new component operation data to evaluate the impact of external environmental factors on component health. Use statistical analysis methods (such as Pearson correlation coefficient) to determine the relationship between noise and component performance. Correlation threshold: 0.7 (when the correlation coefficient is greater than 0.7, it is considered that the noise has a significant impact on the component) Use the current, voltage, power and other data of the equipment to monitor the operating status of the motor in real time and filter out the electrical data directly related to the operation of the component. Current sampling frequency: 1 kHz (1000 data acquisitions per second), voltage sampling frequency: 500 Hz (500 data acquisitions per second). Use Fourier transform (FFT) to perform frequency analysis on the current and voltage signals of the motor to extract the current frequency and voltage frequency characteristics. Current frequency: 0-200 Hz, voltage frequency: 0-200 Hz. Use frequency domain analysis to calculate the amplitude spectrum of current and voltage to evaluate the operating stability of the motor.Based on the results of current and voltage frequency analysis, a health assessment model (such as a threshold-based detection method) is used to determine whether the electrical system is abnormal. Current stability: amplitude change rate ≤5%, voltage stability: amplitude change rate ≤3%. Verify the effectiveness of the electrical health assessment method by comparing the data of normal and faulty motors. Verify the accuracy of the assessment results by comparing the current / voltage frequency data in healthy and faulty states. Combine external changes (such as environmental factors) and internal changes (such as electrical system health) to generate comprehensive health impact data to provide accurate component health predictions. Multi-dimensional data fusion of external change influencing factors (such as environmental noise) and internal change influencing factors (such as electrical system status) is performed, and weighted average or multivariate regression analysis methods are used. External change factor weight: 0.6, internal change factor weight: 0.4. Use a weighted sum model to weight and summarize each factor to generate component health impact data. Health impact data = (external impact factor × weight external) + (internal impact factor × weight internal) According to the value of the health impact data, set different health status levels (such as healthy, warning, fault) for classification. Verify the health impact model after data integration in actual applications. For example, by comparing actual equipment failure data, the accuracy of health impact data in predicting failures can be verified.

[0103] Preferably, step S22 includes the following steps:

[0104] Step S221: collecting operating environment noise of the new component operating data to obtain the operating environment noise data of the new component; performing noise signal time-frequency analysis on the operating environment noise data of the new component to generate operating environment noise signal time-frequency data;

[0105] Step S222: performing a signal peak value change trend analysis on the operating environment noise signal according to the operating environment noise signal time-frequency data to generate signal peak value change trend data, wherein the signal peak value change trend data includes peak value gentle change trend data and peak value fluctuation change trend data;

[0106] Step S223: Separate the new component operation noise from the operating environment noise signal time-frequency data by using the peak gentle change trend data to obtain the new component operation noise; separate the environmental noise from the operating environment noise signal time-frequency data by using the peak fluctuation change trend data to obtain the new component operation noise;

[0107] Step S224: analyzing the impact of component operation noise changes on the new component operation noise according to the environmental separation noise, and generating external change impact factors of the new component.

[0108] In the embodiment of the present invention, a highly sensitive noise sensor (such as an accelerometer, a vibration sensor, or a higher precision microphone array) is arranged at a key position of the offshore operating equipment, especially around the components. Sampling frequency: 1 kHz, 1000 data are collected per second to ensure that the high-frequency components of the environmental noise are captured. Sampling time: 30 minutes, which is sufficient to cover the noise change characteristics under normal production operations. Sampling period: different operating conditions (such as high and low equipment load, temperature change, and operating time) are selected to test the environmental noise. The collected environmental noise signal is converted into time and frequency using the STFT method to extract the frequency components in different time windows. The specific steps include: using a 100ms time window and a 50% overlap between windows to ensure that a finer-grained frequency component is analyzed. Set to 0.1 Hz to ensure that the noise characteristics are not lost in the lower frequency range. Generate a time-frequency diagram with the frequency data in each time window to vividly reflect the frequency characteristics of the noise signal changing over time. Extract the peak points of the environmental noise from the time-frequency diagram. Use the local extreme value method or the absolute peak method to identify the maximum value in each time window, which usually represents the instantaneous intensity of the noise signal. A threshold of the noise signal is set. Fluctuations below this value are not considered valid noise peaks. It is usually set to ±3 times the standard deviation of the signal, that is, only the noise peaks with significant changes are considered, and a peak data point is extracted every 50ms. The peak data is smoothed using the sliding average method, and the average value in each time window is calculated. This can extract the long-term change trend of the noise and identify the more continuous and slowly changing environmental noise (such as continuous changes caused by temperature and humidity). A 100ms window is usually selected for smoothing. The average peak value of ±50ms around each time point is calculated, and the smoothed data obtained is the gentle change trend of the environmental noise. Peak changes with instantaneous fluctuations are identified by calculating the standard deviation of the peak or using fluctuation analysis. For example, the sudden noise change in the factory or the instantaneous vibration when the equipment starts, the change rate of the peak is calculated, and the large fluctuation is marked as peak fluctuation change. Independent component analysis is used to separate the operating noise and environmental noise of the components from the mixed signal. This method assumes that the environmental noise and component noise are statistically independent and attempts to separate the two from the total signal. According to the peak change trend obtained from the previous analysis, use a bandpass filter to filter the signal to remove high-frequency and low-frequency noise and retain the new component operation noise in the mid-frequency band. Use the peak gentle change trend data as the basis for filtering to remove the noise caused by long-term changes outside the equipment. Use the peak fluctuation change trend data to identify sudden environmental noise and separate it from the signal to obtain the new component operation noise. Use regression analysis methods (such as multivariate linear regression and support vector machine regression) to establish a relationship model between the new component operation noise and environmental noise.Assume that environmental noise (external variation factor) will directly affect the operating status of components. Use the time-frequency data of environmental noise and component operating noise as the input of the model, and output the external variation influencing factor. Select features such as peak value, frequency, and duration as input features. Use the training set to train the regression model, and optimize the model parameters through cross-validation.

[0109] Preferably, step S23 includes the following steps:

[0110] Step S231: filtering the equipment motor operation data of the new component operation data to obtain the equipment motor operation data;

[0111] Step S232: extracting current parameters and voltage parameters from the equipment motor operation data to obtain equipment motor operation current parameters and equipment motor operation voltage parameters; performing fast Fourier transform on the equipment motor operation current parameters and equipment motor operation voltage parameters to generate equipment motor operation current spectrum diagram and equipment motor operation voltage spectrum diagram;

[0112] Step S233: performing spectrum analysis on the equipment motor running current spectrum diagram and the equipment motor running voltage spectrum diagram to generate equipment motor current frequency data and equipment motor voltage frequency data; overlapping the equipment motor current frequency data and the equipment motor voltage frequency data to generate an equipment motor running spectrum overlapping curve;

[0113] Step S234: Calculate the spectrum fluctuation mean of the equipment motor operation spectrum overlap curve to obtain the equipment motor operation spectrum mean data; screen the spectrum fluctuation influencing parameters of the equipment motor operation data based on the equipment motor operation spectrum mean data to generate the internal change influencing factors of the new component.

[0114] In an embodiment of the present invention, the current, voltage and related control signals of the motor are extracted from the real-time operation data collected from multi-source sensors (such as current sensors, voltage sensors, speed sensors, etc.). The feature analysis of time domain and frequency domain is used for screening. First, the current and voltage signals in the operation of the motor are screened to eliminate the data related to the non-motor. According to the experimental setting, the current and voltage signals with a sampling frequency of 1kHz are selected to ensure that the dynamic response of the motor is captured. The current and voltage data are extracted from the screened device motor operation data. Taking the current sampling rate of 1kHz as the standard, the current and voltage data for 1 minute are selected for analysis. The amplitude, frequency, phase and other parameters of the motor current and voltage are extracted, which will reflect the working state of the motor under different load conditions. Fast Fourier transform (FFT) is applied to the current and voltage data respectively to convert the time domain data into frequency domain data. The specific steps are: FFT is performed on the current signal to obtain a current spectrum diagram. FFT is performed on the voltage signal to obtain a voltage spectrum diagram. The frequency resolution is set to 0.1 Hz so that the frequency components of the motor operation can be accurately captured. The generated current spectrum and voltage spectrum include the main frequency components of the motor during operation, such as operating frequency, harmonic frequency, etc. The operating frequency of the motor is determined by analyzing the main frequency peak in the current spectrum. This frequency is usually the synchronous frequency of the motor or its multiple. The voltage spectrum is analyzed in the same way to extract the main frequency peak and the frequency components of the higher harmonics of the motor. The peaks of the current spectrum and the voltage spectrum are compared and overlapped to analyze the similarities between the two. Through overlapping spectra, the relationship between the operating frequency and harmonic frequency of the motor can be observed. Using Matlab or Python's fft library, align the data of the current spectrum and the voltage spectrum and generate overlapping curves. Ensure that the spectra of the two are matched on the same time axis and mark the main frequency points. Perform statistical analysis on the overlapping spectrum curve of the motor and calculate the mean of its spectrum fluctuation. The specific method is to calculate the standard deviation and mean in the spectrum curve to quantify the fluctuation of the spectrum. Calculation formula: ;in is the spectrum data point, is the average value of the spectrum, is the total number of data points. Based on the spectrum fluctuation mean, the frequency range that has a greater impact on the motor operation is screened out, especially the impact of low-frequency fluctuations and high-frequency noise on the motor performance. The selected spectrum fluctuation mean is used as an influencing factor to evaluate the health of the motor under different load conditions. This factor can be used to reflect the health changes inside the equipment, such as motor wear and aging.

[0115] As an example of the present invention, refer to Figure 3 As shown, in this example, step S3 includes:

[0116] Step S31: performing fatigue damage analysis on the old component operation data to generate old component fatigue damage data;

[0117] Step S32: allocating state weights to the health impact data of new components and fatigue damage data of old components based on the entropy weight method to generate state weight data of components of offshore equipment;

[0118] Step S33: dividing the offshore equipment component status weight data into a data set to generate a model training set and a model test set; using a convolutional neural network algorithm to perform model training on the model training set to generate a pre-model for predicting the health status of offshore equipment components;

[0119] Step S34: perform model optimization iteration on the pre-model for predicting the health status of offshore operating equipment components through the model test set, thereby generating a health status prediction model for operating components; import the offshore operating equipment component status weight data into the operating component health status prediction model to predict the health status of offshore operating equipment, and generate operating component health status prediction data.

[0120] In the embodiment of the present invention, the operation data of the old type components are collected under different working conditions (such as high load, low load, etc.), mainly including mechanical load, vibration signal, temperature, rotation speed, etc. These signals are analyzed in time domain and frequency domain, especially vibration signals, to identify early signs of fatigue damage. The SN curve (stress-life curve) or the ore method (Palmgren-Miner law) is used to evaluate fatigue damage. The stress level of the component under the working load is measured and combined with historical data to predict the fatigue damage of the component. The average value, peak value, root mean square (RMS) value, etc. are used to evaluate fatigue damage. Spectral analysis is used to detect the occurrence of fatigue cracks by observing the changes in high-frequency components. The generated fatigue damage data includes the stress distribution of the component under different working conditions, the damage evolution trend, and the predicted remaining life. The health impact data of the new component and the fatigue damage data of the old component are standardized to ensure the uniformity of the dimensions of each data. The entropy value of each indicator is calculated, and the entropy value reflects the information uncertainty of the indicator. The richer the information, the smaller the entropy value; the scarcer the information, the larger the entropy value. According to the size of the entropy value, the weight of each indicator is calculated. The smaller the entropy value, the greater the weight. The health impact data of new components (such as environmental change influencing factors, internal change influencing factors) and the fatigue damage data of old components (such as damage index, stress change) are standardized. According to the calculated entropy value, the weight of each factor is obtained, and these weights are assigned to various indicators to reflect their relative importance in the prediction of component health status. The generated offshore equipment component status weight data is used in the subsequent health status prediction model training. The offshore equipment component status weight data is divided into a training set and a test set at a ratio of 80% and 20%. The training set is used for model training, and the test set is used for model evaluation and verification. The divided data is normalized or standardized to ensure the validity and accuracy of the data when it is input into the model. A convolutional neural network is constructed using multiple convolutional layers, pooling layers, and fully connected layers to extract deep features from component health status data. The structure of the network can be designed as follows: The first convolutional layer: filter size 3x3, step size 1, ReLU activation function. Pooling layer: maximum pooling, pooling size is 2x2. The second convolutional layer: filter size 3x3, step size 1, ReLU activation function. Fully connected layer: composed of several neurons, using softmax activation function. The batch gradient descent (SGD) or Adam optimization algorithm is used for model training, and the cross-entropy loss function (Cross-Entropy Loss) is used to evaluate the performance of the model. After the training, a preliminary model (prediction pre-model) for predicting the health status of offshore equipment components is obtained. By using the divided test set, the accuracy and generalization ability of the training model are verified to ensure that the convolutional neural network can effectively predict the health status. The model is back-propagated and error corrected using the test set, and the network weights are continuously adjusted to optimize the prediction results.The learning rate decay, dropout, data enhancement and other techniques are used to improve the accuracy and stability of the model. After each iteration, the model is evaluated by calculating the loss function and accuracy, and optimized according to the evaluation results. The optimized model can be used for real-time health status prediction of offshore equipment. The weight data of the status of offshore equipment components is imported into the model for prediction, and the health status evaluation of each component is generated. The final health status prediction model of the operating component is generated, and this model is used to predict the health status of the components in actual production, and the health status prediction data is output.

[0121] Preferably, step S31 includes the following steps:

[0122] Step S311: performing stress distribution calculation on the old component operation data to generate the old component operation stress distribution data; performing cyclic loading extraction on the old component operation stress distribution data to obtain a cyclic loading data set; performing load frequency statistics on the cyclic loading data set to generate load frequency distribution data;

[0123] Step S312: analyzing the material characteristics of the old-type component on the old-type component operation data to generate the material characteristic data of the old-type component; constructing the SN curve on the material characteristic data of the old-type component to generate the SN curve of the old-type component;

[0124] Step S313: Based on Miner's linear damage accumulation theory, regional damage assessment is performed on the load frequency distribution data and the SN curve of the old component to generate fatigue damage data of the old component.

[0125] In an embodiment of the present invention, stress, vibration, temperature and other data of the old type components during operation are collected by sensors and data acquisition systems. Pay special attention to stress data, and the stress on the surface or key position of the component can be measured by strain gauges or stress sensors. The data should cover the operation conditions under different working conditions, including parameters such as the rotation speed, load, and temperature of the component. For example, under different conditions of high load and low load, the stress distribution of the component will be significantly different. Finite element analysis (FEA) technology is used to calculate the stress distribution of the old type components under different working conditions. By establishing a finite element model of the component, inputting material properties, loads and boundary conditions, the stress values ​​on the surface and inside of each component are calculated, and the stress distribution data of the old type components under different working conditions are obtained. These data can be expressed as stress-strain curves or maximum stress values ​​at different positions. Cyclic loading extraction is performed on the stress distribution data to identify the different load cycles experienced by the component during work. Using statistical methods, load frequency analysis is performed to extract the frequency of occurrence under different loads and generate load frequency distribution data, which can be used to describe the number of times the component experiences different load states throughout its life cycle. Collect mechanical property data of old component materials, including tensile strength, yield strength, ductility, hardness, fatigue strength, etc. Material properties can usually be obtained from material databases, experimental tests or design manuals. In experiments, these performance data can be measured by laboratory material tests (such as tensile tests, fatigue tests, etc.). Use the fatigue limit and durability data of old component materials to construct the SN curve. The SN curve represents the life of the material under different stress levels (that is, the relationship between stress and number of cycles). Use experimental data or SN curve data in standard literature, and use fitting technology to establish an SN curve suitable for the material. The commonly used fitting method is the double logarithmic regression method: ;in, is the fatigue life (number of cycles), is the stress amplitude, and The SN curve of the old component is generated by taking the material characteristic constant as the material characteristic constant, which can provide the durability information of the component under different stresses. According to the ore method theory, the component will produce a certain amount of damage in each cycle during the fatigue process, and the accumulation of damage can be expressed by the following formula: ;in is the total damage degree, For the The number of cycles experienced in a load range, is the fatigue life under this stress range (obtained from the SN curve). For each load frequency range, use the load frequency distribution data and SN curve obtained in the previous step to calculate the damage of each load range. Accumulate the damage of all load ranges to obtain the total fatigue damage degree, and generate the fatigue damage data of the old component, which is expressed as the overall fatigue damage state of the component under the current working environment.

[0126] Preferably, step S4 comprises the following steps:

[0127] Step S41: Visualizing the health status of the operating components based on their health status prediction data to generate a health status display interface for offshore operating equipment;

[0128] Step S42: constructing maintenance decisions for the health status prediction data of the operating components according to the health status display interface of the offshore operating equipment, and generating a health status decision for the offshore operating equipment.

[0129] In the embodiment of the present invention, the health status prediction data of the offshore operation equipment components are generated based on the aforementioned steps (such as step S3), and these data include the health status, remaining life, failure probability, damage degree, etc. of the equipment components.

[0130] The data may include the health score of each component, the operating efficiency of the equipment, vibration analysis data, electrical parameters (such as current, voltage), etc. Display the overall health status of offshore equipment, including four health status levels: good, normal, warning, and fault. The health status of each component is intuitively displayed through colors, icons, etc. The health status level can be divided according to the health score of the component (such as health index or remaining life). For example: good: health score > 80%, normal: health score 50%-80%, warning: health score 30%-50%, fault: health score <30%. Display the probability of failure and remaining life of each component, using bar charts, pie charts, etc. Display the health trend of the equipment through a dashboard or line chart to reflect the changing trend of the health status of the equipment in real time. Choose appropriate visualization technology and platform for display. Common implementation methods include: using HTML, CSS, JavaScript (such as D3.js, Chart.js and other libraries) to achieve dynamic display of health status. Use graphical interface design tools (such as Qt, C#, etc.) to achieve local display of health status. Use data streaming platforms (such as Apache Kafka) combined with front-end frameworks (such as React, Vue) to provide real-time data updates and displays. Obtain health status prediction data of offshore equipment components from the data acquisition system and transmit it to the front-end display system through API or message queue. In the front-end interface, map the data with specific visualization elements (such as charts, colors, indicator lights, etc.). For example, the health status score is represented by color changes, and the value is displayed through a dashboard or bar chart. Ensure that the health status display interface can update data in real time and provide dynamic health status information. Update the display content regularly through WebSocket or timer mechanism to generate the offshore equipment health status display interface, including: health status, failure probability, remaining life of each component, overall health trend chart, and real-time health data dynamic display. Set the threshold of the health score to distinguish different health status levels. Set the data update frequency (such as once a minute or once an hour) to ensure that the displayed health status information is the latest. Based on the health status data obtained in the health status display interface, formulate maintenance decision rules. Generally, decision rules can be constructed based on the health status, failure probability and remaining life of equipment components. Good health status: The component continues to operate normally and no intervention is required. Normal health status: If the component health status score drops, it needs to be checked during the scheduled maintenance window. Health status warning: If the component health status score is below a certain threshold, it is recommended to conduct an emergency inspection or arrange maintenance. Health status failure: When a component fails or is seriously damaged, it needs to be shut down and repaired immediately. Build a decision model based on the health status prediction data. The decision model can be established through weighted average method, fuzzy logic, decision tree or machine learning algorithm.For example, historical data can be used for regression analysis or classification model training to predict the maintenance needs of various components. Based on decision rules and decision variables, the system automatically generates maintenance decisions. For example, based on real-time updates of health status data of offshore equipment, the system automatically recommends whether preventive maintenance, component replacement, or emergency shutdown is required. Decision tree algorithms or heuristic methods can be used to optimize maintenance plans. For example, for components with low health scores, the system can issue warnings and recommend scheduling inspections. For components with a higher probability of failure, the system recommends emergency overhauls.

[0131] In this specification, a system for evaluating the health status of offshore equipment is provided, which is used to execute the above-mentioned method for evaluating the health status of offshore equipment. The system for evaluating the health status of offshore equipment includes:

[0132] The data acquisition module is used to obtain the data of the components of the offshore operation equipment; to perform initial life marking on the data of the components of the offshore operation equipment to generate the initial life marking data of the components of the offshore operation equipment; to perform real-time operation data collection on the data of the components of the offshore operation equipment to generate the real-time operation data of the components of the offshore operation equipment;

[0133] The new equipment analysis module is used to divide the real-time operation data of offshore equipment components based on the components with initial life mark data, and generate the operation data of new components and old components; perform environmental change impact analysis on the operation data of new components, and generate the external change impact factors of new components; perform electrical system health assessment on the operation data of new components, and generate the internal change impact factors of new components; integrate the external change impact factors of new components and the internal change impact factors of new components to generate the health impact data of new components;

[0134] The old equipment analysis module is used to perform fatigue damage analysis on the operation data of old components and generate fatigue damage data of old components; predict the health status of offshore operating equipment based on the health impact data of new components and fatigue damage data of old components and generate health status prediction data of operating components;

[0135] The health decision module is used to build maintenance decisions based on the health status prediction data of operating components and generate health status decisions for offshore operating equipment.

[0136] The beneficial effect of the present invention is that by collecting the operating data of the components of offshore equipment in real time, it is possible to ensure dynamic monitoring of the equipment status, timely discover potential faults or performance degradation, and reduce the risk of unexpected equipment downtime. By marking the component data with initial life and dividing it into new components and old components, it is helpful to adopt differentiated maintenance strategies for components with different life cycles, which not only improves the accuracy of maintenance, but also enables targeted management based on the actual conditions of the components. Environmental change impact analysis and electrical system health assessment of new components can identify the impact of external environment and internal changes of components on equipment health. This multi-dimensional analysis method helps to comprehensively evaluate the health status of the equipment and then formulate more accurate prediction and maintenance strategies. Fatigue damage analysis of old components can identify potential damage accumulation areas in advance and help predict the remaining service life of components. Predicting the health status of offshore equipment in combination with the health impact data of new components can improve the accuracy of fault warnings and reduce downtime and maintenance costs caused by equipment failures. Maintenance decision-making based on equipment health status prediction data can optimize maintenance plans and ensure that repairs or parts are replaced when the equipment is in optimal working condition, thereby reducing unnecessary downtime and excessive maintenance and improving overall production efficiency and equipment utilization. Through accurate health status prediction and maintenance decisions, the service life of equipment can be extended, the risk of unexpected downtime in the production process can be reduced, and it also helps to control maintenance costs and improve resource utilization efficiency. Therefore, the present invention improves the accuracy of health status assessment of offshore equipment through multi-dimensional data collection, environmental and electrical impact analysis, fatigue damage assessment and intelligent prediction.

[0137] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0138] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for evaluating the health status of offshore equipment, characterized in that: The following steps are involved: Step S1: Acquire offshore operation equipment component data; Perform initial life marking on offshore equipment component data to generate initial life marking data for offshore equipment components; perform real-time operation data collection on offshore equipment component data to generate real-time operation data for offshore equipment components; Step S2: Based on the components of the initial life mark data, the real-time operation data of the offshore equipment components are divided into components to generate the operation data of new components and the operation data of old components; the environmental change impact analysis is performed on the operation data of the new components to generate the external change impact factors of the new components; the electrical system health assessment is performed on the operation data of the new components to generate the internal change impact factors of the new components; the external change impact factors of the new components and the internal change impact factors of the new components are integrated to generate the health impact data of the new components; Step S2 includes the following steps: Step S21: dividing the real-time operation data of the offshore equipment components based on the components of the initial life mark data to generate new component operation data and old component operation data; Step S22: collecting operating environment noise of the new component operating data to obtain operating environment noise data of the new component; performing noise separation on the operating environment noise data of the new component to generate operating noise of the new component and environmental separation noise; performing environmental change impact analysis on the operating noise of the new component according to the environmental separation noise to generate external change impact factors of the new component; Step S23: Screening the operation data of the new component motor to obtain the operation data of the new component motor; performing current and voltage frequency analysis on the operation data of the new component motor to generate the current frequency data of the new component motor and the voltage frequency data of the new component motor; performing electrical system health assessment on the current frequency data of the new component motor and the voltage frequency data of the new component motor to generate the internal change influencing factor of the new component motor; Step S24: integrating the external change influencing factors of the new component and the internal change influencing factors of the new component to generate health impact data of the new component; Step S3: Perform fatigue damage analysis on the operation data of the old type components to generate fatigue damage data of the old type components; perform health status prediction of the offshore operation equipment according to the health impact data of the new type components and the fatigue damage data of the old type components to generate health status prediction data of the operation components; Step S3 includes the following steps: Step S31: performing fatigue damage analysis on the old component operation data to generate old component fatigue damage data; Step S32: allocating state weights to the health impact data of new components and fatigue damage data of old components based on the entropy weight method to generate state weight data of components of offshore equipment; Step S33: dividing the offshore equipment component status weight data into a data set to generate a model training set and a model test set; using a convolutional neural network algorithm to perform model training on the model training set to generate a pre-model for predicting the health status of offshore equipment components; Step S34: optimizing and iterating the pre-model for predicting the health status of offshore operating equipment components through the model test set, thereby generating a health status prediction model for operating components; importing the state weight data of offshore operating equipment components into the health status prediction model for operating components to predict the health status of offshore operating equipment, thereby generating health status prediction data for operating components; Step S4: construct maintenance decisions based on the health status prediction data of the operating components to generate health status decisions for offshore operating equipment.

2. The method for evaluating the health status of offshore equipment according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: Acquire offshore operation equipment component data; Step S12: preprocessing the offshore equipment component data to generate standard offshore equipment component data, wherein the data preprocessing includes data cleaning, data denoising, data missing value filling and data standardization; Step S13: Perform component life cycle analysis on the standard offshore equipment component data to generate component life cycle data; perform initial life marking on the standard offshore equipment component data based on the component life cycle data to generate initial life marking data for the offshore equipment component; Step S14: using multi-source sensors to collect operating data of standard offshore operating equipment components, and updating the collected operating data in real time based on a preset time interval to generate real-time operating data of offshore operating equipment components.

3. The method for evaluating the health status of offshore equipment according to claim 2, characterized in that: The full life cycle analysis of standard offshore equipment component data includes: Conduct demand analysis and modeling on the data of standard offshore equipment components to generate functional demand model data; conduct process design simulation on the functional demand model data to generate component design phase data; Optimize the processing flow of the component design phase data to generate processing flow optimization data; perform quality control inspection on the processing flow optimization data to generate quality inspection report data; perform assembly accuracy analysis on the quality inspection report data to generate component production phase data; Conduct operation monitoring and analysis on component production phase data to generate health status assessment data; conduct fault diagnosis analysis on health status assessment data to generate fault diagnosis report data; perform performance optimization and adjustment on fault diagnosis report data to generate component use phase data; Perform periodic maintenance analysis on component use phase data to generate maintenance plan data; perform repair effect verification on maintenance plan data to generate component maintenance phase data; perform retirement assessment analysis on component maintenance phase data to generate retirement assessment data; perform disassembly process analysis on retirement assessment data to generate component retirement phase data; Based on the data from the component design stage, component production stage, component usage stage, component maintenance stage and component retirement stage, the data of the entire life cycle of equipment operation is integrated to generate component full life cycle data.

4. The method for evaluating the health status of offshore equipment according to claim 1, characterized in that: Step S22 includes the following steps: Step S221: collecting operating environment noise of the new component operating data to obtain the operating environment noise data of the new component; performing noise signal time-frequency analysis on the operating environment noise data of the new component to generate operating environment noise signal time-frequency data; Step S222: performing a signal peak value change trend analysis on the operating environment noise signal according to the operating environment noise signal time-frequency data to generate signal peak value change trend data, wherein the signal peak value change trend data includes peak value gentle change trend data and peak value fluctuation change trend data; Step S223: Separate the new component operation noise from the operating environment noise signal time-frequency data by using the peak gentle change trend data to obtain the new component operation noise; separate the environmental noise from the operating environment noise signal time-frequency data by using the peak fluctuation change trend data to obtain the new component operation noise; Step S224: analyzing the impact of component operation noise changes on the new component operation noise according to the environmental separation noise, and generating external change impact factors of the new component.

5. The method for evaluating the health status of offshore equipment according to claim 1, characterized in that: Step S23 includes the following steps: Step S231: filtering the equipment motor operation data of the new component operation data to obtain the equipment motor operation data; Step S232: extracting current parameters and voltage parameters from the equipment motor operation data to obtain equipment motor operation current parameters and equipment motor operation voltage parameters; performing fast Fourier transform on the equipment motor operation current parameters and equipment motor operation voltage parameters to generate equipment motor operation current spectrum diagram and equipment motor operation voltage spectrum diagram; Step S233: performing spectrum analysis on the equipment motor running current spectrum diagram and the equipment motor running voltage spectrum diagram to generate equipment motor current frequency data and equipment motor voltage frequency data; overlapping the equipment motor current frequency data and the equipment motor voltage frequency data to generate an equipment motor running spectrum overlapping curve; Step S234: Calculate the spectrum fluctuation mean of the equipment motor operation spectrum overlap curve to obtain the equipment motor operation spectrum mean data; screen the spectrum fluctuation influencing parameters of the equipment motor operation data based on the equipment motor operation spectrum mean data to generate the internal change influencing factors of the new component.

6. The method for evaluating the health status of offshore equipment according to claim 1, characterized in that: Step S31 includes the following steps: Step S311: performing stress distribution calculation on the old component operation data to generate the old component operation stress distribution data; performing cyclic loading extraction on the old component operation stress distribution data to obtain a cyclic loading data set; performing load frequency statistics on the cyclic loading data set to generate load frequency distribution data; Step S312: analyzing the material characteristics of the old-type component on the old-type component operation data to generate the material characteristic data of the old-type component; constructing the SN curve on the material characteristic data of the old-type component to generate the SN curve of the old-type component; Step S313: Based on Miner's linear damage accumulation theory, regional damage assessment is performed on the load frequency distribution data and the SN curve of the old component to generate fatigue damage data of the old component.

7. The method for evaluating the health status of offshore equipment according to claim 1, characterized in that: Step S4 includes the following steps: Step S41: Visualizing the health status of the operating components based on their health status prediction data to generate a health status display interface for offshore operating equipment; Step S42: constructing maintenance decisions for the health status prediction data of the operating components according to the health status display interface of the offshore operating equipment, and generating a health status decision for the offshore operating equipment.

8. A system for evaluating the health status of offshore equipment, characterized in that: The method for evaluating the health status of offshore equipment according to claim 1 is used to perform the evaluation system for the health status of offshore equipment, the evaluation system comprising: The data acquisition module is used to obtain the data of the components of the offshore operating equipment; to perform initial life marking on the data of the components of the offshore operating equipment to generate the initial life marking data of the components of the offshore operating equipment; to perform real-time operation data collection on the data of the components of the offshore operating equipment to generate the real-time operation data of the components of the offshore operating equipment; The new equipment analysis module is used to divide the real-time operation data of offshore equipment components based on the components with initial life mark data, and generate the operation data of new components and old components; perform environmental change impact analysis on the operation data of new components, and generate the external change impact factors of new components; perform electrical system health assessment on the operation data of new components, and generate the internal change impact factors of new components; integrate the external change impact factors of new components and the internal change impact factors of new components to generate the health impact data of new components; The old equipment analysis module is used to perform fatigue damage analysis on the operation data of old components and generate fatigue damage data of old components; predict the health status of offshore operating equipment based on the health impact data of new components and fatigue damage data of old components and generate health status prediction data of operating components; The health decision module is used to build maintenance decisions based on the health status prediction data of operating components and generate health status decisions for offshore operating equipment.

Images