Subsea gate valve condition monitoring analysis method and system

Abstract

The application discloses a seabed gate valve state monitoring analysis method and system, relates to the field of gate valve state monitoring, and comprises the following steps: a collecting module is used for collecting valve opening and closing action data, medium pressure and flow data and ocean temperature and salinity data in a preset range around the valve during seabed gate valve operation to form an initial monitoring data set; a calibration module is used for receiving the initial monitoring data set, performing ocean environment interference stripping, equipment coupling noise elimination and space-time dimension correlation calibration processing on the initial monitoring data set, and generating standardized data suitable for feature extraction; and the application constructs a performance degradation model by using key parameters related to sealing and valve rods to scientifically evaluate the remaining useful life and generate a message containing basic, abnormal and life information.

Description

Submarine gate valve condition monitoring and analysis method and system Technical Field

[0001] This invention relates to the field of gate valve condition monitoring technology, specifically to a method and system for monitoring and analyzing the condition of subsea gate valves. Background Technology

[0002] Subsea gate valves are core fluid control components in subsea oil and gas extraction and transportation systems. Their valve bodies are mostly made of super duplex steel or nickel-based alloys that are resistant to seawater corrosion, resisting marine electrochemical corrosion and biological adhesion. The sealing structure adopts a combination of metal hard seal and elastic seal to achieve zero leakage under high pressure of 10-150MPa, and is suitable for water depth environments of 100-3000 meters. The operating end integrates an underwater hydraulic or electric actuator, which receives ground control signals through an umbilical cable to complete remote opening and closing. It also has an emergency shutdown function to ensure the fluid control safety and stable operation of the subsea oil and gas transportation system.

[0003] Currently, for the daily operation and maintenance of subsea gate valves, most existing technologies are equipped with simple sensors to perceive real-time data for maintenance, and are equipped with real-time control of maintenance cycles. However, the timeliness and accuracy of the monitoring of the subsea gate valve's status are poor.

[0004] To address this, we propose a method and system for monitoring and analyzing the condition of subsea gate valves. Summary of the Invention

[0005] In view of the above-mentioned shortcomings of the existing technology, the present invention provides a method and system for monitoring and analyzing the status of submarine gate valves, which can effectively solve the problems of the existing technology.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions;

[0007] This invention discloses a condition monitoring and analysis system for subsea gate valves, comprising:

[0008] The system comprises the following modules: Acquisition module, Calibration module, and Extraction module. Acquisition module collects data on valve opening and closing actions, medium pressure and flow rate, and ocean temperature and salinity within a preset range to form an initial monitoring dataset. Calibration module receives the initial monitoring dataset and performs marine environmental interference removal, equipment coupling noise elimination, and spatiotemporal correlation calibration to generate standardized data suitable for feature extraction. Extraction module receives the standardized data and extracts feature parameters reflecting the valve's sealing state and stem wear to construct a state feature dataset. Identification module establishes a mapping model between feature parameters in the state feature dataset and the valve's normal operating state. Based on the model, it identifies whether real-time feature parameters deviate from the normal range; if so, it outputs an anomaly identification result. Evaluation module applies the anomaly identification result and the state feature dataset to construct a valve performance degradation model and assess the valve's remaining usable life. Generation module receives the anomaly identification result and the valve's remaining life assessment result, generates and stores a valve safety control message based on both.

[0009] The acquisition module is interconnected with the calibration module via a wireless network. The calibration module is interconnected with the extraction module via a wireless network. The extraction module is interconnected with the identification module and the evaluation module via a wireless network. The identification module and the evaluation module are interconnected with the generation module via a wireless network.

[0010] Furthermore, the acquisition module includes a multi-dimensional sensing unit and a data synchronization unit;

[0011] The multi-dimensional sensing unit includes a displacement sensor deployed at the valve stem drive end of the gate valve, a pressure sensor and a flow sensor deployed at the inlet and outlet ends of the gate valve, and a temperature sensor and a salinity sensor deployed on the outer wall of the gate valve and within a preset range around it.

[0012] The data synchronization unit synchronizes data based on timestamps, aligning the data collected by each sensor with the same time granularity, thus establishing the correlation between valve opening and closing action data, medium pressure and flow data, and ocean temperature and salinity data in the time dimension to form an initial monitoring dataset.

[0013] Furthermore, during the marine environmental disturbance stripping operation, an environmental disturbance correction model is constructed based on marine temperature and salinity data to calculate the impact of environmental disturbances on the medium's pressure and flow rate data.

[0014] ;

[0015] In the formula: The pressure data of the medium after environmental interference has been removed; The raw medium pressure data collected; This is the coefficient representing the effect of temperature on pressure. This is measured ocean temperature data; The standard temperature under the design operating conditions of the gate valve; This is the coefficient representing the effect of salinity on pressure; This is based on measured ocean salinity data; Standard salinity under gate valve design operating conditions;

[0016] When the equipment coupling noise elimination operation is executed, an adaptive filtering algorithm is used. The vibration data of the gate valve during no-load operation is used as the reference noise to filter out the noise from the pressure and flow data and valve opening and closing action data after the marine environment interference stripping stage.

[0017] When performing spatiotemporal dimension correlation calibration, the pressure and flow data, temperature and salinity data and valve action status within the same time interval are correlated and marked based on the time series of the gate valve opening and closing actions, and finally standardized data are generated.

[0018] Furthermore, when extracting characteristic parameters reflecting the sealing status of the gate valve and the degree of stem wear, the extraction module follows the following rules:

[0019] Sealing condition characteristic parameters include the pressure difference fluctuation coefficient of the sealing surface. ;

[0020] In the formula: Number of samples; This represents the pressure difference data across the gate valve sealing surface collected for the i-th time. The average pressure difference from n samples; This refers to the number of times the differential pressure fluctuation exceeds a preset threshold within the sampling period;

[0021] The characteristic parameters of valve stem wear are characterized by calculating the valve stem stroke deviation rate and the valve stem driving torque change rate. The valve stem stroke deviation rate is the percentage of the difference between the actual opening and closing stroke and the theoretical design stroke to the theoretical design stroke. The valve stem driving torque change rate is the percentage of the difference between the average driving torque in the current operating cycle and the average driving torque in the initial operating cycle to the initial average driving torque.

[0022] The sealing surface pressure difference fluctuation coefficient, valve stem stroke deviation rate, and valve stem driving torque change rate are integrated to construct a state feature dataset.

[0023] Furthermore, when constructing the feature parameter-normal operation status mapping model in the identification module, normal operation data of the gate valve at different health stages within its design life cycle are collected to build a feature parameter sample library.

[0024] A dynamic mapping model between feature parameters and normal operating status is constructed based on a sample database. The model uses the cumulative operating time t of the gate valve as a variable and outputs the normal threshold range of each feature parameter for the corresponding time. The model expression is:

[0025] ;

[0026] In the formula: Let be the lower and upper limits of the normal threshold for the feature parameters at time t; The normal baseline value of the characteristic parameter at time t; The threshold correction coefficient at time t; The overall normal fluctuation standard deviation of the characteristic parameter;

[0027] During anomaly detection, feature parameters are extracted in real time. The dynamic threshold interval corresponding to time t Compare the features extracted in real time. consecutive exceed If the duration reaches the preset duration, or exceeds the limit in a single instance... If the preset amplitude is reached, the gate valve is determined to be abnormal, and the abnormality identification result is output, including the type of abnormal characteristic parameters. Duration of the abnormality.

[0028] Furthermore, the operations in the evaluation module for constructing a gate valve performance degradation model and assessing the remaining usable life are as follows:

[0029] Using the sealing surface pressure difference fluctuation coefficient, valve stem stroke deviation rate, and valve stem driving torque change rate from the state feature dataset as input variables, and the correspondence between feature parameters and actual remaining life in the historical operating data of gate valves as training samples, a performance degradation model is constructed:

[0030] ;

[0031] In the formula: Remaining usable lifetime; This refers to the pressure difference fluctuation coefficient at the sealing surface. Valve stem stroke deviation rate; This represents the rate of change of the valve stem driving torque. , , , , Here are the model coefficients, where Base lifetime correction;

[0032] Input the sealing surface pressure difference fluctuation coefficient, valve stem stroke deviation rate, and valve stem driving torque change rate from the real-time status feature dataset into the performance degradation model, and output the current remaining usable life of the gate valve.

[0033] Furthermore, security control messages include basic information segments, anomaly information segments, and lifetime information segments;

[0034] The basic information section includes the gate valve's unique identifier code, monitoring timestamp, and the operating status of the acquisition and calibration modules; the anomaly information section includes the anomaly type, anomaly characteristic parameter values, and anomaly duration in the anomaly identification results; and the lifespan information section includes the remaining usable lifespan assessment results.

[0035] When storing security control messages, the messages are associated with the unique identifier of the gate valve and stored together. At the same time, a message index is created, with index dimensions including monitoring time, anomaly type, and remaining lifespan range. This allows for quick querying of historical messages by index.

[0036] Furthermore, during the evaluation module's operation phase, historical anomaly identification results output by the identification module are received synchronously, and the sampling frequency of the acquisition module and the correction coefficient of the calibration module are dynamically adjusted:

[0037] If the remaining usable lifetime is lower than the preset lifetime threshold, or the historical anomaly frequency is higher than the preset frequency threshold, the sampling frequency of the acquisition module is increased, and the coefficients of the environmental interference correction model in the calibration module are optimized; if the remaining usable lifetime is higher than the preset lifetime threshold and the historical anomaly frequency is lower than the preset frequency threshold, the sampling frequency of the acquisition module is decreased.

[0038] The optimization logic for the coefficients of the environmental disturbance correction model is as follows:

[0039] ;

[0040] In the formula: The optimized temperature influence coefficient and salinity influence coefficient; The temperature influence coefficient and salinity influence coefficient before optimization; For adaptive learning rate, ∈ (0,1); To correct for pressure errors, the difference between the medium pressure data after removing environmental interference and the theoretical pressure value of the gate valve under standard operating conditions is taken.

[0041] On the other hand, the methods for monitoring and analyzing the condition of subsea gate valves include:

[0042] Data on valve opening and closing actions, medium pressure and flow rate, and surrounding ocean temperature and salinity are collected and an initial monitoring dataset is constructed based on timestamp alignment. An environmental interference correction model is built based on the ocean temperature and salinity data from the initial monitoring data to remove environmental interference from the medium pressure and flow rate data, simultaneously eliminating equipment coupling noise, and generating standardized data adapted for feature extraction. Feature parameters reflecting the sealing status and stem wear of the subsea gate valve are extracted from the standardized data, and a state feature dataset is constructed based on these extracted parameters. Normal operation data at different health stages throughout the gate valve's lifespan are collected, and a dynamic mapping model of feature parameters and normal operation status is constructed. Real-time feature parameters are compared with the dynamic threshold range output by the model to determine and output anomaly identification results. Using the sealing surface pressure difference fluctuation coefficient, stem stroke deviation rate, and driving torque change rate from the state feature dataset as inputs, a performance degradation model is constructed based on the gate valve's historical operating data to assess the remaining usable life. Simultaneously, the sampling frequency and calibration correction coefficients are adjusted according to the remaining life and historical anomaly frequency. Based on the anomaly identification results and the remaining usable life assessment results, a safety control message containing basic information, anomaly information, and lifespan information is generated. This message is associated with the gate valve's unique identifier and stored, and an index is established.

[0043] Compared with the known prior art, the technical solution provided by this invention has the following beneficial effects:

[0044] This invention provides a method and system for monitoring and analyzing the status of a subsea gate valve. During execution, the method and system can comprehensively collect data on the opening and closing actions, medium pressure and flow rate, and surrounding ocean temperature and salinity during the operation of the gate valve. Data synchronization is achieved by using timestamps. An environmental interference correction model is constructed to remove the influence of the environment on the pressure and flow data. Adaptive filtering is used to eliminate equipment coupling noise. Spatiotemporal dimension correlation calibration is performed simultaneously to improve the standardization of data. A dynamic threshold model is constructed based on normal data from different health stages throughout the gate valve's life cycle. The threshold range can be adjusted according to the operating time to accurately identify anomalies and output the anomaly type, parameter difference, and duration.

[0045] Meanwhile, a performance degradation model is constructed by using key parameters related to sealing and valve stem to scientifically assess the remaining usable life. Messages containing basic, abnormal, and life information are generated and stored in association. Multi-dimensional indexes are established for easy querying. The sampling frequency and correction coefficient can also be dynamically adjusted based on the remaining life and the frequency of historical anomalies to improve monitoring accuracy and efficiency, effectively ensure the long-term reliable operation of the subsea gate valve, and reduce failure risks and maintenance costs. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0047] Figure 1 is a schematic diagram of the underwater gate valve condition monitoring and analysis system;

[0048] Figure 2 is a flowchart illustrating the condition monitoring and analysis method for subsea gate valves. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0050] The present invention will be further described below with reference to embodiments.

[0051] Example 1:

[0052] The subsea gate valve status monitoring and analysis system of this embodiment, as shown in Figure 1, includes:

[0053] The data acquisition module is used to collect data on valve opening and closing actions, medium pressure and flow rate, and ocean temperature and salinity within a preset range during the operation of the subsea gate valve, forming an initial monitoring dataset.

[0054] The acquisition module includes a multi-dimensional sensing unit and a data synchronization unit;

[0055] The multi-dimensional sensing unit includes a displacement sensor deployed at the valve stem drive end of the gate valve, a pressure sensor and a flow sensor deployed at the inlet and outlet ends of the gate valve, and a temperature sensor and a salinity sensor deployed on the outer wall of the gate valve and within a preset range around it.

[0056] The data synchronization unit synchronizes data based on timestamps, aligning the data collected by each sensor with the same time granularity, so as to make the valve opening and closing action data, medium pressure and flow data and ocean temperature and salinity data correlated in the time dimension to form an initial monitoring dataset.

[0057] Among them, the displacement sensor is used to collect the stroke, speed and duration data of the valve opening and closing action, the pressure sensor and flow sensor are used to collect medium pressure data and flow data, and the temperature sensor and salinity sensor are used to collect ocean temperature data and salinity data.

[0058] The calibration module is used to receive the initial monitoring dataset, perform marine environmental interference removal, equipment coupling noise elimination, and spatiotemporal dimension correlation calibration on the initial monitoring dataset, and generate standardized data that is adapted for feature extraction.

[0059] During the marine environmental disturbance stripping operation, an environmental disturbance correction model is constructed based on marine temperature and salinity data to calculate the impact of environmental disturbances on the medium's pressure and flow rate data.

[0060] ;

[0061] In the formula: The pressure data of the medium after environmental interference has been removed; The raw medium pressure data collected; This is the coefficient representing the effect of temperature on pressure. This is measured ocean temperature data; The standard temperature under the design operating conditions of the gate valve; This is the coefficient representing the effect of salinity on pressure; This is based on measured ocean salinity data; Standard salinity under gate valve design operating conditions;

[0062] The above formula achieves environmental interference removal by quantifying the impact of ocean temperature and salinity on the original medium pressure data. First, using the standard temperature and salinity under the gate valve design conditions as a benchmark, the deviations of the measured temperature and salinity from the standard values ​​are calculated respectively. Then, combined with the temperature influence coefficient determined by the gate valve material and medium characteristics, and the salinity influence coefficient determined by the correlation characteristics between medium density and seawater salinity, the pressure influence corresponding to the two types of deviations is subtracted from the original pressure data to obtain the corrected pressure data. This design not only considers the two key interference factors of temperature and salinity in the marine environment, but also makes the correction process adaptable to gate valves with different materials and different medium characteristics through the differential setting of coefficients, effectively removing environmental interference, improving the accuracy of pressure data, and laying a reliable data foundation for subsequent feature extraction.

[0063] in, , All values ​​are greater than zero. The value is determined by the material of the gate valve and the characteristics of the medium. The higher the thermal expansion coefficient of the gate valve material and the stronger the thermal compressibility of the medium, the larger the value. The lower the thermal expansion coefficient of the gate valve material and the weaker the thermal compressibility of the medium, the smaller the value. The value is determined by the correlation between medium density and seawater salinity. The higher the correlation between medium density and seawater salinity and the greater the density of the medium itself, the larger the value. Conversely, the lower the correlation between medium density and seawater salinity and the smaller the density of the medium itself, the smaller the value.

[0064] When the equipment coupling noise elimination operation is executed, an adaptive filtering algorithm is used. The vibration data of the gate valve during no-load operation is used as the reference noise. The pressure and flow data and valve opening and closing action data after the marine environment interference stripping stage are used to filter out noise in order to eliminate the coupling noise generated by valve stem friction and shell vibration during equipment operation.

[0065] When performing the spatiotemporal dimension correlation calibration operation, based on the time series of the gate valve opening and closing actions, the pressure flow data, temperature and salinity data within the same time interval are correlated and marked with the valve action status, and finally standardized data is generated.

[0066] The valve's operating states include: open, closed, and pressure holding;

[0067] The extraction module is used to receive standardized data, extract feature parameters reflecting the sealing status and stem wear of the gate valve based on the standardized data, and construct a status feature dataset.

[0068] When extracting characteristic parameters reflecting the sealing condition and stem wear of the gate valve, the extraction module follows the following rules:

[0069] Sealing condition characteristic parameters include the pressure difference fluctuation coefficient of the sealing surface. ;

[0070] In the formula: Number of samples; This represents the pressure difference data across the gate valve sealing surface collected for the i-th time. The average pressure difference from n samples; This refers to the number of times the differential pressure fluctuation exceeds a preset threshold within the sampling period;

[0071] The above formula calculates the absolute value of the deviation between multiple sets of measured differential pressure and average differential pressure, and then combines the differential pressure fluctuation frequency to obtain the differential pressure fluctuation coefficient of the sealing surface in the form of an average value. This allows for a more comprehensive quantification of the pressure stability of the sealing surface. This design differs from the limitation of judging the sealing status solely by differential pressure deviation or fluctuation frequency. By combining the two, it can more accurately capture subtle changes in the sealing surface and provide key indicators for identifying abnormal gate valve sealing.

[0072] The characteristic parameters of valve stem wear are characterized by calculating the valve stem stroke deviation rate and the valve stem driving torque change rate. The valve stem stroke deviation rate is the percentage of the difference between the actual opening and closing stroke and the theoretical design stroke to the theoretical design stroke. The valve stem driving torque change rate is the percentage of the difference between the average driving torque in the current operating cycle and the average driving torque in the initial operating cycle to the initial average driving torque.

[0073] The sealing surface pressure difference fluctuation coefficient, valve stem stroke deviation rate, and valve stem driving torque change rate are integrated to construct a state feature dataset.

[0074] The identification module is used to establish a mapping relationship model between the feature parameters in the state feature dataset and the feature parameters of the gate valve's normal operating state. Based on the model, it identifies whether the real-time feature parameters deviate from the normal range. If they deviate, it outputs the anomaly identification result.

[0075] When constructing the feature parameter-normal operation status mapping model in the identification module, normal operation data of the gate valve at different health stages within the design life cycle are collected to build a feature parameter sample library.

[0076] A dynamic mapping model between feature parameters and normal operating status is constructed based on a sample database. The model uses the cumulative operating time t of the gate valve as a variable and outputs the normal threshold range of each feature parameter for the corresponding time. The model expression is:

[0077] ;

[0078] In the formula: Let be the lower and upper limits of the normal threshold for the feature parameters at time t; The normal baseline value of the characteristic parameter at time t; The threshold correction coefficient at time t; The overall normal fluctuation standard deviation of the characteristic parameter is calculated from normal samples throughout the entire life cycle.

[0079] The above formula achieves effective monitoring of the gate valve's operating status by dynamically setting the normal range boundaries of real-time characteristic parameters. Based on the baseline values ​​of characteristic parameters under normal operating conditions, the lower and upper limits of the normal range are calculated by combining the coefficients and standard deviations that change over time. The dynamic adjustment of the coefficients can adapt to the characteristic changes of the gate valve at different operating stages, while the standard deviation takes into account the reasonable fluctuation range of the characteristic parameters. This ensures that the normal range boundaries not only conform to the actual operating rules of the gate valve but also flexibly respond to subtle changes during operation, making anomaly identification more targeted and accurate, and providing a basis for timely detection of gate valve operating anomalies.

[0080] >0, determined by the dispersion of normal characteristic parameters near time t in the sample database; the greater the dispersion, the better. A larger value ensures that the normal fluctuation range is covered;

[0081] The above model enables dynamic adjustment of the threshold range as the gate valve operates for a period of time, in order to adapt to the normal performance degradation process of the gate valve.

[0082] During anomaly detection, feature parameters are extracted in real time. The dynamic threshold interval corresponding to time t Compare the features extracted in real time. consecutive exceed If the duration reaches the preset duration, or exceeds the limit in a single instance... If the preset amplitude is reached, the gate valve is determined to be abnormal, and the abnormality identification result is output, including the type of abnormal characteristic parameters. Duration of the anomaly;

[0083] The evaluation module is used to apply anomaly identification results and state feature datasets to build a gate valve performance degradation model and evaluate the remaining usable life of the gate valve.

[0084] The operations for building a gate valve performance degradation model and evaluating remaining usable life in the evaluation module are as follows:

[0085] Using the sealing surface pressure difference fluctuation coefficient, valve stem stroke deviation rate, and valve stem driving torque change rate from the state feature dataset as input variables, and the correspondence between feature parameters and actual remaining life in the historical operating data of gate valves as training samples, a performance degradation model is constructed:

[0086] ;

[0087] In the formula: Remaining usable lifetime; This refers to the pressure difference fluctuation coefficient at the sealing surface. Valve stem stroke deviation rate; This represents the rate of change of the valve stem driving torque. , , , , Here are the model coefficients, where Base lifetime correction;

[0088] The above formula takes the key characteristic parameters affecting the performance of gate valves as the core to construct the remaining usable life calculation model. The sealing surface pressure difference fluctuation coefficient, valve stem stroke deviation rate, and valve stem driving torque change rate are used as input variables. By setting different coefficients, the influence of each parameter on the remaining life is quantified. Then, combined with the basic life correction term, the final remaining usable life is obtained. The coefficients are assigned different values ​​according to the sensitivity of the gate valve to different performance, component characteristics, and initial health state. For example, the coefficients of gate valves with high sealing performance sensitivity are larger, so that the model can be adapted to different types of gate valves under different working conditions, making the remaining usable life assessment more in line with the actual situation.

[0089] in, >0 indicates that the value is larger when the gate valve is highly sensitive to sealing performance (e.g., when conveying high pressure or corrosive media, sealing failure directly leads to serious operating risks), and smaller when the gate valve is low sensitive to sealing performance (e.g., when conveying low pressure or non-corrosive media, slight deterioration of the seal does not significantly affect operation). <0 indicates that the value is larger (the absolute value is smaller and closer to 0) when the gate valve sealing material has strong wear resistance and anti-aging ability (the rate of life decay caused by seal deterioration is slow). The value is smaller (the absolute value is larger and further away from 0) when the gate valve sealing material is easy to wear and age (the rate of life decay caused by seal deterioration is fast). <0 indicates that when the gate valve stem guide structure is stable and the stroke deviation has a weak impact on the core function (such as large-diameter gate valves, where small stroke deviations do not hinder the flow of media), the value is larger (the absolute value is smaller and closer to 0). When the gate valve has high requirements for the accuracy of the valve stem stroke (such as precision control conditions, where stroke deviations can easily lead to incomplete opening and closing), the value is smaller (the absolute value is larger and further away from 0). <0 indicates that when the gate valve drive mechanism has high load redundancy and strong torque fluctuation tolerance (such as a high-power drive motor, where slight torque changes do not damage the transmission components), the value is larger (the absolute value is smaller and closer to 0). When the gate valve drive mechanism has weak load capacity and is sensitive to torque changes (such as a low-power drive device, where abnormal torque can easily cause transmission failure), the value is smaller (the absolute value is larger and further away from 0). >0 indicates that the value is larger when the gate valve has a long initial design life, high historical maintenance frequency, and excellent maintenance quality (good basic health status and high remaining life benchmark when the characteristic parameters are optimal), and smaller when the gate valve has a short initial design life, insufficient historical maintenance, or existing faults (poor basic health status and low remaining life benchmark when the characteristic parameters are optimal).

[0090] Input the sealing surface pressure differential fluctuation coefficient, valve stem stroke deviation rate, and valve stem driving torque change rate from the real-time status feature dataset into the performance degradation model, and output the current remaining usable life of the gate valve.

[0091] The generation module is used to receive the anomaly identification results and the gate valve remaining life assessment results, generate gate valve safety control messages based on the two and store them;

[0092] Security control messages include basic information segments, anomaly information segments, and lifespan information segments;

[0093] The basic information section includes the gate valve's unique identifier code, monitoring timestamp, and the operating status of the acquisition and calibration modules; the anomaly information section includes the anomaly type, anomaly characteristic parameter values, and anomaly duration in the anomaly identification results; and the lifespan information section includes the remaining usable lifespan assessment results.

[0094] When storing security control messages, the security control messages are associated with the unique identifier of the gate valve and stored together. At the same time, a message index is established, and the index dimensions include monitoring time, anomaly type, and remaining lifespan range. It supports quick query of historical messages by index.

[0095] During the evaluation module's operation phase, it synchronously receives historical anomaly identification results output by the identification module and dynamically adjusts the sampling frequency of the acquisition module and the correction coefficient of the calibration module.

[0096] If the remaining usable lifetime is lower than the preset lifetime threshold, or the historical anomaly frequency is higher than the preset frequency threshold, the sampling frequency of the acquisition module is increased, and the coefficients of the environmental interference correction model in the calibration module are optimized; if the remaining usable lifetime is higher than the preset lifetime threshold and the historical anomaly frequency is lower than the preset frequency threshold, the sampling frequency of the acquisition module is decreased.

[0097] The optimization logic for the coefficients of the environmental disturbance correction model is as follows:

[0098] ;

[0099] In the formula: The optimized temperature influence coefficient and salinity influence coefficient; The temperature influence coefficient and salinity influence coefficient before optimization; For adaptive learning rate, ∈ (0,1); To correct for pressure error, the difference between the medium pressure data after removing environmental interference and the theoretical pressure value of the gate valve under standard operating conditions is taken.

[0100] The above formula improves the model's accuracy in correcting environmental disturbances by dynamically optimizing the coefficients of the environmental disturbance correction model. Based on the temperature and salinity influence coefficients before optimization, the optimized coefficients are calculated by combining the adaptive learning rate and the partial derivatives of the pressure correction error with respect to the coefficients. Based on the above formula, the coefficients of the environmental disturbance correction model can be dynamically adjusted according to the actual monitoring data, continuously improving the correction effect on environmental disturbances and avoiding the problem of decreased correction accuracy caused by fixed coefficients. This provides more accurate standardized data for subsequent data processing.

[0101] These are the partial derivatives of the pressure correction error with respect to the temperature influence coefficient and the salinity influence coefficient, respectively. The value of follows: The larger the value, the larger the value. The smaller the value, the smaller the value.

[0102] The acquisition module is interconnected with the calibration module via a wireless network. The calibration module is interconnected with the extraction module via a wireless network. The extraction module is interconnected with the identification module and the evaluation module via a wireless network. The identification module and the evaluation module are interconnected with the generation module via a wireless network.

[0103] In this embodiment, the acquisition module collects data on the valve opening and closing actions, medium pressure and flow rate, and ocean temperature and salinity within a preset range during the operation of the subsea gate valve, forming an initial monitoring dataset. The calibration module then receives the initial monitoring dataset and performs marine environmental interference removal, equipment coupling noise elimination, and spatiotemporal dimension correlation calibration on the initial monitoring dataset to generate standardized data suitable for feature extraction. The extraction module then receives the standardized data and extracts feature parameters reflecting the gate valve's sealing status and valve stem wear based on the standardized data, constructing a state feature dataset. The identification module further establishes a mapping relationship model between the feature parameters in the state feature dataset and the feature parameters of the gate valve's normal operating state. Based on the model, it identifies whether the real-time feature parameters deviate from the normal range. If they deviate, it outputs an anomaly identification result. The evaluation module applies the anomaly identification result and the state feature dataset to construct a gate valve performance degradation model to evaluate the remaining usable life of the gate valve. Finally, the generation module receives the anomaly identification result and the gate valve's remaining life evaluation result, generates a gate valve safety control message based on both, and stores it.

[0104] In the above embodiments, when the system is applied to the status monitoring and analysis of subsea gate valves, it can accurately collect relevant data on the operation of subsea gate valves and the surrounding marine environment, eliminate environmental interference and equipment noise, accurately grasp the sealing status of the gate valve and the wear of the valve stem, promptly identify operational anomalies and clarify abnormal information, reliably evaluate the remaining usable life, and at the same time, adjust the sampling frequency and correction parameters according to the life and abnormal conditions to improve monitoring accuracy, ensure the safe operation of the gate valve, reduce failures, extend the service life, and facilitate historical data query and maintenance.

[0105] Example 2:

[0106] At the implementation level, based on Example 1, this example further describes the submarine gate valve status monitoring and analysis system in Example 1 with reference to Figure 2:

[0107] Methods for monitoring and analyzing the condition of subsea gate valves include:

[0108] Collect data on valve opening and closing actions, medium pressure and flow rate, and surrounding ocean temperature and salinity, and construct an initial monitoring dataset based on timestamp alignment;

[0109] An environmental interference correction model is constructed based on ocean temperature and salinity data from the initial monitoring data to remove environmental interference from medium pressure and flow data, simultaneously eliminate equipment coupling noise, and generate standardized data adapted to feature extraction.

[0110] Feature parameters reflecting the sealing status and stem wear of the subsea gate valve are extracted from standardized data, and a status feature dataset is constructed based on the extracted feature parameters.

[0111] Collect normal operation data of gate valves at different health stages throughout their entire life cycle, construct a dynamic mapping model of feature parameters and normal operation status, compare real-time feature parameters with the dynamic threshold range output by the model, and determine and output anomaly identification results.

[0112] The sealing surface pressure difference fluctuation coefficient, valve stem stroke deviation rate and driving torque change rate in the state feature dataset are used as inputs. A performance degradation model is constructed by combining the historical operating data of the gate valve to evaluate the remaining usable life. At the same time, the sampling frequency and calibration correction coefficient are adjusted according to the remaining life and the frequency of historical anomalies.

[0113] Based on the anomaly identification results and the remaining usable life assessment results, a safety control message containing basic information, anomaly information, and lifespan information is generated. The message is associated with the gate valve's unique identification code and stored, and an index is established.

[0114] In summary, the methods and systems described in the above embodiments can comprehensively collect data on the opening and closing actions of the gate valve, medium pressure and flow rate, and surrounding ocean temperature and salinity during operation. Data synchronization is achieved using timestamps. An environmental interference correction model is constructed to eliminate the influence of the environment on pressure and flow data. Adaptive filtering is used to eliminate equipment coupling noise. Spatiotemporal correlation calibration is performed simultaneously to improve data standardization. A dynamic threshold model is constructed based on normal data from different health stages throughout the gate valve's lifespan. This model can adjust the threshold range according to the operating time, accurately identify anomalies, and output the anomaly type, parameter difference, and duration. Simultaneously, a performance degradation model is constructed using key parameters related to sealing and valve stem to scientifically assess the remaining usable lifespan. Messages containing basic, anomaly, and lifespan information are generated and stored in association. A multi-dimensional index is established for easy querying. Furthermore, the sampling frequency and correction coefficients can be dynamically adjusted based on the remaining lifespan and historical anomaly frequency, improving monitoring accuracy and efficiency, effectively ensuring the long-term reliable operation of the subsea gate valve, and reducing failure risks and maintenance costs.

[0115] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A condition monitoring and analysis system for subsea gate valves, characterized in that, include: The data acquisition module is used to collect data on valve opening and closing actions, medium pressure and flow rate, and ocean temperature and salinity within a preset range during the operation of the subsea gate valve, forming an initial monitoring dataset. The calibration module is used to receive the initial monitoring dataset, perform marine environmental interference removal, equipment coupling noise elimination, and spatiotemporal dimension correlation calibration on the initial monitoring dataset, and generate standardized data that is adapted for feature extraction. The extraction module is used to receive standardized data, extract feature parameters reflecting the sealing status and stem wear of the gate valve based on the standardized data, and construct a status feature dataset. The identification module is used to establish a mapping relationship model between the feature parameters in the state feature dataset and the feature parameters of the gate valve's normal operating state. Based on the model, it identifies whether the real-time feature parameters deviate from the normal range. If they deviate, it outputs the anomaly identification result. The evaluation module is used to construct a gate valve performance degradation model and evaluate the remaining usable life of the gate valve by applying the anomaly identification results and the state feature dataset. The operation of constructing the gate valve performance degradation model and evaluating the remaining usable life in the evaluation module is as follows: using the sealing surface pressure difference fluctuation coefficient, valve stem stroke deviation rate, and valve stem driving torque change rate from the state feature dataset as input variables, and using the correspondence between feature parameters and actual remaining life in the gate valve's historical operating data as training samples, a performance degradation model is constructed. In the formula: Remaining usable lifetime; This refers to the pressure difference fluctuation coefficient at the sealing surface. Valve stem stroke deviation rate; This represents the rate of change of the valve stem driving torque. 、 、 、 、 Here are the model coefficients, where The basic lifespan correction term is used; the sealing surface pressure difference fluctuation coefficient, valve stem stroke deviation rate, and valve stem driving torque change rate from the real-time state feature dataset are input into the performance degradation model, and the current remaining usable lifespan of the gate valve is output. The generation module is used to receive the anomaly identification results and the gate valve remaining life assessment results, and generate and store the gate valve safety control message based on the two.

2. The subsea gate valve status monitoring and analysis system according to claim 1, characterized in that, The acquisition module includes a multi-dimensional sensing unit and a data synchronization unit. The multi-dimensional sensing unit includes a displacement sensor deployed at the valve stem drive end of the gate valve, a pressure sensor and a flow sensor deployed at the inlet and outlet ends of the gate valve, and a temperature sensor and a salinity sensor deployed on the outer wall of the gate valve and within a preset range around it. The data synchronization unit synchronizes the data based on timestamps, aligning the data collected by each sensor with the same time granularity, so that the valve opening and closing action data, medium pressure and flow data, and ocean temperature and salinity data are correlated in the time dimension to form the initial monitoring dataset.

3. The subsea gate valve status monitoring and analysis system according to claim 1, characterized in that, During the execution of the marine environmental disturbance stripping operation, an environmental disturbance correction model is constructed based on marine temperature and salinity data to calculate the impact of environmental disturbances on the medium's pressure and flow rate data. In the formula: The pressure data of the medium after environmental interference has been removed; The raw medium pressure data collected; This is the coefficient representing the effect of temperature on pressure. This is measured ocean temperature data; The standard temperature under the design operating conditions of the gate valve; This is the coefficient representing the effect of salinity on pressure; This is based on measured ocean salinity data; The standard salinity is set for the gate valve's design operating conditions. During the execution of the equipment coupling noise elimination operation, an adaptive filtering algorithm is used, taking the vibration data of the gate valve during no-load operation as reference noise, to filter out noise from the pressure and flow data and valve opening and closing action data after the marine environment interference stripping stage. During the execution of the spatiotemporal dimension correlation calibration operation, based on the time series of the gate valve's opening and closing actions, pressure and flow data, temperature and salinity data within the same time interval are correlated and marked with the valve's action status, ultimately generating the standardized data.

4. The subsea gate valve status monitoring and analysis system according to claim 1, characterized in that, When extracting characteristic parameters reflecting the sealing state and stem wear of the gate valve, the extraction module follows the following principle: sealing state characteristic parameters include the pressure difference fluctuation coefficient of the sealing surface. In the formula: Number of samples; This represents the pressure difference data across the gate valve sealing surface collected for the i-th time. The average pressure difference from n samples; This refers to the number of times the differential pressure fluctuation exceeds a preset threshold within the sampling period; The characteristic parameters of valve stem wear are characterized by calculating the valve stem stroke deviation rate and the valve stem driving torque change rate. The valve stem stroke deviation rate is the percentage of the difference between the actual opening and closing stroke and the theoretical design stroke to the theoretical design stroke. The valve stem driving torque change rate is the percentage of the difference between the average driving torque in the current operating cycle and the average driving torque in the initial operating cycle to the initial average driving torque. The sealing surface pressure difference fluctuation coefficient, valve stem stroke deviation rate, and valve stem driving torque change rate are integrated to construct a state characteristic dataset.

5. The subsea gate valve status monitoring and analysis system according to claim 1, characterized in that, When constructing the feature parameter-normal operation status mapping model in the identification module, the normal operation data of the gate valve at different health stages within the design life cycle are collected to build a feature parameter sample library. A dynamic mapping model between feature parameters and normal operating status is constructed based on a sample database. The model uses the cumulative operating time t of the gate valve as a variable and outputs the normal threshold range of each feature parameter for the corresponding time. The model expression is: In the formula: Let be the lower and upper limits of the normal threshold for the feature parameters at time t; The normal baseline value of the characteristic parameter at time t; The threshold correction coefficient at time t; The standard deviation of the overall normal fluctuation of the feature parameters; during anomaly identification, the feature parameters extracted in real time will be used. The dynamic threshold interval corresponding to time t Compare the features extracted in real time. consecutive exceed If the duration reaches the preset duration, or exceeds the limit in a single instance... If the preset amplitude is reached, the gate valve is determined to be abnormal, and the abnormality identification result is output, including the type of abnormal characteristic parameters. Duration of the abnormality.

6. The subsea gate valve status monitoring and analysis system according to claim 1, characterized in that, The security control message includes a basic information segment, an anomaly information segment, and a lifespan information segment; the basic information segment includes the gate valve's unique identifier code, monitoring timestamp, and the operating status of the acquisition module and calibration module; the anomaly information segment includes the anomaly type, anomaly characteristic parameter values, and anomaly duration in the anomaly identification results; The lifespan information segment includes the remaining usable lifespan assessment results; when storing the safety control message, the safety control message is associated with the gate valve's unique identifier code and stored together, and a message index is established. The index dimensions include monitoring time, anomaly type, and remaining lifespan range, supporting quick query of historical messages by index.

7. The subsea gate valve status monitoring and analysis system according to claim 1, characterized in that, During the operation phase of the evaluation module, it synchronously receives historical anomaly identification results output by the identification module and dynamically adjusts the sampling frequency of the acquisition module and the correction coefficients of the calibration module: if the remaining usable lifetime is lower than a preset lifetime threshold, or the frequency of historical anomalies is higher than a preset frequency threshold, the sampling frequency of the acquisition module is increased, and the coefficients of the environmental interference correction model in the calibration module are optimized; if the remaining usable lifetime is higher than a preset lifetime threshold and the frequency of historical anomalies is lower than a preset frequency threshold, the sampling frequency of the acquisition module is decreased; the optimization logic of the coefficients of the environmental interference correction model is as follows: In the formula: The optimized temperature influence coefficient and salinity influence coefficient; The temperature influence coefficient and salinity influence coefficient before optimization; For adaptive learning rate, ∈（0,1）； To correct for pressure errors, the difference between the medium pressure data after removing environmental interference and the theoretical pressure value of the gate valve under standard operating conditions is taken.

8. The subsea gate valve status monitoring and analysis system according to claim 1, characterized in that, The acquisition module is interconnected with the calibration module via a wireless network. The calibration module is interconnected with the extraction module via a wireless network. The extraction module is interconnected with the identification module and the evaluation module via a wireless network. The identification module and the evaluation module are interconnected with the generation module via a wireless network.

9. A method for monitoring and analyzing the condition of a subsea gate valve, wherein the method is an implementation method of the subsea gate valve condition monitoring and analysis system as described in any one of claims 1-8, characterized in that, include: Collect data on valve opening and closing actions, medium pressure and flow rate, and surrounding ocean temperature and salinity, and construct an initial monitoring dataset based on timestamp alignment; An environmental interference correction model is constructed based on ocean temperature and salinity data from the initial monitoring data to remove environmental interference from medium pressure and flow data, simultaneously eliminating equipment coupling noise, and generating standardized data adapted for feature extraction. Feature parameters reflecting the sealing status and stem wear of the subsea gate valve are extracted from the standardized data, and a state feature dataset is constructed based on the extracted feature parameters. Normal operation data of different health stages throughout the life cycle of the gate valve are collected, and a dynamic mapping model of feature parameters-normal operation status is constructed. The real-time feature parameters are compared with the dynamic threshold range output by the model to determine and output the anomaly identification results. Using the sealing surface pressure difference fluctuation coefficient, valve stem stroke deviation rate, and driving torque change rate from the state feature dataset as inputs, a performance degradation model is constructed based on the historical operating data of the gate valve to evaluate the remaining usable life. At the same time, the sampling frequency and calibration correction coefficient are adjusted according to the remaining life and the frequency of historical anomalies. Based on the anomaly identification results and the remaining usable life evaluation results, a safety control message containing basic information, anomaly information, and life information is generated. The message is associated with the gate valve's unique identification code and stored, and an index is established.

Images