A blowout preventer internal ram stroke position real-time monitoring and calibration method

Abstract

The application discloses a blowout preventer internal gate stroke position real-time monitoring and calibration method, comprising the following steps: synchronously collecting hydraulic pressure sequence, hydraulic flow sequence, valve control instruction sequence, action time sequence and acoustic emission event stream when the gate opening and closing action is triggered; performing pretreatment on the collected data; constructing pressure-flow-instruction response trajectory; extracting segmented slope feature, hysteresis shape feature, response time delay feature and inter-segment transition feature, and performing position inversion processing to obtain gate stroke position preliminary estimation sequence; identifying true-to-position endpoint moment, generating stroke zero point anchor point and stroke range endpoint anchor point; performing event parameterization processing on the pretreated acoustic emission event stream; and performing stroke calibration processing. The stroke inversion calibration method of the application utilizes multi-source standardized sequence and acoustic emission anchor point constraint, realizes gate position real-time monitoring and true-to-position determination, and has the advantages of strong anti-interference, low misjudgment rate and traceable calibration.

Description

Technical Field

[0001] This invention relates to the field of monitoring and calibration, and in particular to a method for real-time monitoring and calibration of the travel position of the internal gate of a blowout preventer. Background Technology

[0002] Current monitoring of blowout preventer (BOP) gate opening and closing status primarily relies on purely mechanical physical displays. These displays may show the gate's open / closed state via indicator rod positions, mechanical stroke scales, or external mechanical indicators. Operators then visually assess and manually record the status, using the visual confirmation as a reference for interlocking controls. However, under conditions of high pressure, strong vibration, mud contamination, and temperature drift in well control systems, mechanical displays are susceptible to assembly gaps, frictional hysteresis, and springback. Visual assessments are also prone to obstruction and differences in experience, resulting in insufficient consistency in opening and closing status confirmation and hindering the formation of continuous, quantitative monitoring results for the gate's stroke position. Summary of the Invention

[0003] One objective of this invention is to propose a method for real-time monitoring and calibration of the travel position of the internal gate of a blowout preventer. This invention utilizes a travel inversion calibration method based on multi-source standardized sequences and acoustic emission anchor point constraints to achieve real-time monitoring and true positioning determination of the gate position, and has the advantages of strong anti-interference, low false judgment rate and traceable calibration.

[0004] A method for real-time monitoring and calibration of the stroke position of the internal gate of a blowout preventer according to an embodiment of the present invention includes the following steps:

[0005] When the gate opening and closing action is triggered, the hydraulic pressure sequence, hydraulic flow sequence, valve control command sequence, action timing sequence and acoustic emission event stream are collected simultaneously;

[0006] The collected data is preprocessed to generate a set of multi-source standardized sequences of gate action.

[0007] Based on the multi-source standardized sequence set of gate action, a pressure-flow-command response trajectory is constructed and divided into the starting segment, constant speed segment, contact segment, clamping segment and arrival segment;

[0008] Extract segmented slope features, hysteresis morphology features, response delay features, and inter-segment transition features from the starting segment, constant speed segment, contact segment, pressing segment, and final position segment to generate a set of motion digital fingerprint vectors. Then, perform position inversion processing to obtain a preliminary estimation sequence of the gate travel position.

[0009] Within the candidate time window for the gate opening and closing action, identify the true arrival endpoint time and generate the zero point anchor point and the range endpoint anchor point of the stroke.

[0010] Perform event parameterization processing on the preprocessed acoustic emission event stream to identify the key moments of the gate contact section and the gate pressing section, and map them as the contact section position constraint set and the pressing section position constraint set;

[0011] Based on the zero-point anchor point of the travel range, the endpoint anchor point of the travel range, the set of position constraints of the contact section, the set of position constraints of the pressing section, and the preliminary estimation sequence of the gate travel position, the travel calibration process is performed to generate the real-time monitoring results of the gate travel position, the judgment results of the true position of the gate, and the update record of the calibration parameters.

[0012] Optionally, the hydraulic pressure sequence is a pressure data sequence formed by arranging the pressure sampling values ​​of the gate hydraulic execution circuit according to the sampling time during the gate opening and closing operation; the hydraulic flow sequence is a flow data sequence formed by arranging the flow sampling values ​​of the gate hydraulic execution circuit according to the sampling time during the gate opening and closing operation; the valve control command sequence is a command data sequence formed by arranging the valve control opening command, valve control reversing command, and valve control holding command output by the control system according to the command issuance time during the gate opening and closing operation; the action timing sequence is a timing event sequence representing the gate action start time, valve control command issuance time, valve group state switching time, and gate action end time during the gate opening and closing operation; and the acoustic emission event stream is an event data stream formed by arranging the acoustic emission events obtained by collecting acoustic emission signals and triggering detection during the gate opening and closing operation according to the event occurrence time. The acoustic emission event includes the acoustic emission event occurrence time timestamp, acoustic emission event energy, and acoustic emission event count rate.

[0013] Optionally, the generation of the multi-source standardized sequence set of gate action includes:

[0014] The start and end times of the gate action are determined based on the action timing sequence. The time window from the start to the end of the gate action is used as the gate action time window. Data segments corresponding to the hydraulic pressure sequence, hydraulic flow sequence, valve control command sequence and acoustic emission event stream are extracted within the gate action time window.

[0015] Using the start time of the gate action as a unified time reference, the time axes of the hydraulic pressure sequence, hydraulic flow sequence and valve control command sequence are uniformly shifted, and the timestamps of the occurrence time of acoustic emission events in the acoustic emission event stream are uniformly shifted to complete the alignment of the multi-source time axes.

[0016] Anomaly removal and pulsation suppression are performed on the hydraulic pressure sequence and hydraulic flow sequence, and command jitter merging and holding interval merging are performed on the valve control command sequence to form a valve control command change trajectory consistent with the gate action time window;

[0017] Delete acoustic emission events outside the gate action time window and organize the time sequence to obtain the acoustic emission event stream within the gate action time window;

[0018] The processed hydraulic pressure sequence, hydraulic flow sequence, and valve control command change trajectory are merged to generate a set of multi-source standardized sequences of gate action.

[0019] Optionally, the division into the starting section, constant speed section, contact section, pressing section, and final position section specifically includes:

[0020] Obtain a multi-source standardized sequence set of gate action to obtain the hydraulic pressure sequence, hydraulic flow sequence, and valve control command change trajectory within the gate action time window;

[0021] According to the sampling time sequence, the hydraulic pressure sampling value, hydraulic flow sampling value and valve control command change trajectory corresponding to the command state at the same sampling time are combined into a triplet sequence to generate the pressure-flow-command response trajectory.

[0022] Locate the inflection point when the pressure enters the continuously rising range from the initial stable range in the pressure-flow-command response trajectory, determine the end point of the starting segment, and mark the trajectory segment from the start of the gate action to the end point of the starting segment as the starting segment;

[0023] After the end of the starting section, locate the trajectory segment where the hydraulic flow enters the stable platform range and the valve control command changes in the same control state. Mark this trajectory segment as the uniform speed segment, and determine the turning point when the hydraulic flow leaves the stable platform range as the end of the uniform speed segment.

[0024] After the end of the uniform speed segment, locate the moment when the hydraulic pressure suddenly increases and the hydraulic flow simultaneously decreases, mark the trajectory segment from the end of the uniform speed segment to the moment of trajectory turning as the contact segment, and determine the moment of trajectory turning as the end of the contact segment.

[0025] After the end of the contact section, locate the starting moment when the hydraulic pressure enters the high-level slow change range and the valve control command change trajectory enters the holding state. Mark the trajectory segment from the end of the contact section to the starting moment as the clamping section, and determine the starting moment as the end of the clamping section.

[0026] The trajectory segment from the end of the pressing section to the end of the gate action is marked as the positioning section, and the segment boundary times of the starting section, constant speed section, contact section, pressing section and positioning section are written into the pressure-flow-command response trajectory.

[0027] Optionally, the generation of the preliminary estimation sequence of the gate travel position specifically includes:

[0028] Extract the trajectory segments corresponding to the starting segment, constant speed segment, contact segment, clamping segment and arrival segment, and obtain the hydraulic pressure sequence, hydraulic flow sequence and valve control command change trajectory within each trajectory segment;

[0029] Calculate the slope characteristics of the starting segment for the trajectory segment and the response delay characteristics of the starting segment;

[0030] Calculate the segmented slope characteristics and hysteresis morphology characteristics of the uniform velocity segment for the trajectory segment of the uniform velocity segment, and calculate the response delay characteristics of the uniform velocity segment.

[0031] Calculate the contact segment slope characteristics and inter-segment transition characteristics for the contact segment trajectory segment, and calculate the contact segment response delay characteristics;

[0032] The hysteresis morphology and inter-segment transition characteristics of the compaction segment are calculated for the trajectory segment of the compaction segment, and the response delay characteristics of the compaction segment are also calculated.

[0033] Calculate the segment slope characteristics and inter-segment transition characteristics of the trajectory segments in the arrival segment;

[0034] According to the sequence of starting segment, constant speed segment, contact segment, pressing segment and arrival segment, the segment slope characteristics, hysteresis morphology characteristics, response delay characteristics and inter-segment transition characteristics are spliced ​​together to form a set of motion digital fingerprint vectors.

[0035] Position inversion is performed based on the set of motion digital fingerprint vectors to generate a preliminary estimation sequence of the gate travel position.

[0036] Optionally, the execution location inversion specifically includes:

[0037] The motion digital fingerprint vector set is extracted according to the starting segment, constant speed segment, contact segment, pressing segment and arrival segment to obtain five feature groups, and then normalized to obtain normalized five feature groups.

[0038] Perform feature extraction and normalization on each historical motion digital fingerprint vector in the motion digital fingerprint database according to the starting segment, constant speed segment, contact segment, pressing segment and positioning segment to obtain a historical normalized five-segment feature group.

[0039] For each segment, calculate the feature difference and perform a weighted sum to obtain the segment consistency score. Sum the five segment consistency scores to obtain the total segment consistency score set.

[0040] Select the target historical motion digital fingerprint vector corresponding to the minimum segment consistency total score, and read the gate travel position reference sequence corresponding to the target historical motion digital fingerprint vector;

[0041] The gate travel position reference sequence is resampled by linear interpolation according to the sampling time of the gate action time window to generate a preliminary estimation sequence of the gate travel position.

[0042] Optionally, the generation of the zero-point anchor point and the end-point anchor point of the travel range specifically includes:

[0043] Obtain the multi-source standardized sequence set of gate action and the pressure-flow-command response trajectory with completed segment marking, read the start time of the arrival segment and the end time of the gate action, and use the start time of the arrival segment to the end time of the gate action to determine the candidate time window for arrival;

[0044] Within the candidate time window, the pressure waveform segment corresponding to the hydraulic pressure sequence is extracted and high-frequency extraction processing is performed to obtain the high-frequency response waveform of hydraulic pressure. The flow waveform segment corresponding to the hydraulic flow sequence is extracted and transient extraction processing is performed to obtain the transient response waveform of hydraulic flow.

[0045] For each sampling point of the high-frequency response waveform of hydraulic pressure, the adjacent difference is calculated, a sorted sequence of sampling point pairs is generated, and the sampling point pair with the largest absolute value of the difference is selected. The time of the next sampling point of the sampling point pair is defined as the pressure change time, and the pressure change amplitude is generated.

[0046] A ringing analysis window of a preset length is extracted with the moment of pressure change as the center. The set of adjacent peak moments of the high-frequency response waveform of hydraulic pressure within the ringing analysis window is located, and the ringing frequency characteristics are calculated.

[0047] Within the ringing analysis window, the ratio of the first peak amplitude to the last peak amplitude is calculated to obtain the attenuation ratio, and the attenuation duration is generated to obtain the ringing attenuation characteristics.

[0048] The difference between the minimum value of the transient response waveform of hydraulic flow and the steady-state baseline value is calculated in the neighborhood of the moment of pressure change, and the flow rate drop amplitude is generated.

[0049] Within the neighborhood of the pressure change moment, extract the pressure change amplitude, ringing frequency features, ringing attenuation features, and flow rate decline amplitude, and normalize them to construct the arrival transient feature vector. Calculate the vector distance between the arrival transient feature vector and the preset true arrival transient template vector. If the vector distance exceeds the preset distance threshold, replace the pressure change moment with the next sampling point moment corresponding to the next index in the sampling point difference sorting sequence and repeat the above steps until the vector distance does not exceed the preset distance threshold.

[0050] The moment of pressure change that satisfies the vector distance value not exceeding the preset distance threshold is determined as the true end point moment. The gate opening direction or gate closing direction is determined according to the command status identifier of the valve control command change trajectory. The true end point moment corresponding to the gate closing direction is mapped as the stroke zero point anchor point. The true end point moment corresponding to the gate opening direction is mapped as the stroke range end point anchor point.

[0051] Optionally, the generation of the contact segment position constraint set and the clamping segment position constraint set specifically includes:

[0052] The acoustic emission event stream is extracted from the set of multi-source standardized sequences of gate action, and the acoustic emission event stream is sorted according to the timestamp of the occurrence time of the acoustic emission event to form a time-ordered acoustic emission event sequence.

[0053] Perform event parameterization processing on the time-ordered acoustic emission event sequence, representing each acoustic emission event as an acoustic emission event parameter consisting of the timestamp of the occurrence time of the acoustic emission event, the energy of the acoustic emission event, and the count rate of the acoustic emission event, and sum them up to form an acoustic emission event parameter set;

[0054] Statistical analysis is performed on the set of acoustic emission event parameters to generate an acoustic emission event count rate sequence and an acoustic emission event energy sequence. The acoustic emission event count rate sequence and the acoustic emission event energy sequence are then aligned by timestamps to form a joint acoustic emission response sequence.

[0055] Within the contact time range of the pressure-flow-command response trajectory, calculate the energy increment and count rate increment of adjacent time windows for the acoustic emission joint response sequence, locate the timestamp of the acoustic emission event when the energy increment and count rate increment simultaneously reach local peaks, and generate a set of key moments of the gate contact segment.

[0056] Within the time range of the pressure-flow-command response trajectory, the overlapping period of the energy accumulation rise interval and the count rate sustained high interval is calculated for the acoustic emission joint response sequence. The timestamp of the acoustic emission event corresponding to the energy increment reaching the local peak within the overlapping period is located, and a set of key moments of the gate clamping section is generated.

[0057] The key moment set of the gate contact section and the key moment set of the gate pressing section are merged to form a contact anchor point set. The contact anchor point set is then mapped to the corresponding sampling position of the preliminary estimation sequence of the gate stroke position according to the sampling time of the gate action time window, thereby generating a contact section position constraint set and a pressing section position constraint set.

[0058] Optionally, the execution of the stroke calibration process specifically includes:

[0059] Obtain the preliminary estimation sequence of the gate travel position, the zero point anchor point or the end point anchor point of the travel range, the contact section position constraint set and the clamping section position constraint set, and expand the preliminary estimation sequence of the gate travel position according to the sampling time to form a position time series array;

[0060] If a zero-point anchor point exists, the position value corresponding to the sampling time of the zero-point anchor point is read, the difference between the position value and the zero-point position value is calculated as the translation correction amount, and the translation correction amount is synchronously applied to all position values ​​in the position time sequence array to obtain the anchor point aligned position time sequence array;

[0061] If there is a travel range endpoint anchor point, read the position value corresponding to the sampling time of the travel range endpoint anchor point, calculate the ratio of the range endpoint position value to the position value as the scaling correction amount, and apply the scaling correction amount synchronously to all position values ​​in the position time sequence array to obtain the anchor point aligned position time sequence array.

[0062] Based on the anchor point alignment position time sequence array, constraint projection processing is performed on the contact section position constraint set and the clamping section position constraint set. The position value corresponding to the sampling time of each constraint value is replaced with the constraint position value. Piecewise linear interpolation is performed in the preset neighborhood before and after the constraint sampling time to generate a constraint consistent position sequence.

[0063] The constrained consistent position sequence is used as the real-time monitoring result of the gate travel position. Based on the travel zero point anchor point or travel range endpoint anchor point corresponding to the true arrival endpoint, the gate true arrival state judgment result is generated. At the same time, the translation correction amount or scaling correction amount is recorded, and the calibration parameter update record is generated.

[0064] Optionally, the zero point position value is the zero point calibration value of the gate travel position, and the range endpoint position value is the maximum travel calibration value of the gate travel position.

[0065] The beneficial effects of this invention are:

[0066] This invention simultaneously acquires hydraulic pressure sequences, hydraulic flow sequences, valve control command sequences, action timing sequences, and acoustic emission event streams when the gate opening and closing action is triggered. It then performs time alignment and noise suppression preprocessing on the multi-source data to construct a multi-source standardized sequence set of gate actions. This allows key response information throughout the entire gate action process to be continuously expressed under a unified time reference. Based on this multi-source standardized sequence set, a pressure-flow-command response trajectory is constructed and divided into starting, constant speed, contact, clamping, and positioning segments. Furthermore, segment slope features, hysteresis morphology features, response delay features, and inter-segment transition features are extracted from each segment to form a motion digital fingerprint vector set, which is then used for position inversion. This enables the output of a stable preliminary estimate sequence of the gate stroke position under well control conditions where sensor noise, hydraulic pulsation, and valve control disturbances coexist. This avoids position jumps or estimation distortions caused by instantaneous anomalies in the traditional single-point threshold method, thereby improving the continuity and consistency of real-time monitoring of the gate stroke position.

[0067] Meanwhile, within the candidate time window of arrival, this invention extracts the amplitude of pressure change, ringing frequency characteristics, ringing attenuation characteristics, and flow rate decline amplitude by using the high-frequency response waveform of hydraulic pressure and the transient response waveform of hydraulic flow. It constructs a transient feature vector of arrival and performs vector distance verification with a preset true arrival transient template vector. This identifies the true arrival endpoint moment and generates the stroke zero point anchor point and the stroke range endpoint anchor point, so that the endpoint determination has repeatable transient evidence to support it, reducing the risk of false arrival misjudgment caused by valve control command holding, pressure platform drift, or flow fluctuation.

[0068] Furthermore, this invention performs event parameterization processing on the acoustic emission event stream. Within the time range of the contact and clamping sections, it identifies the key moment sets of the gate contact section and the key moment sets of the gate clamping section based on the relationship between energy increment, count rate increment, and local peak values. These sets are then mapped to form the contact section position constraint set and the clamping section position constraint set. This ensures that the stroke calibration process not only relies on the hydraulic circuit response but also introduces independent observational evidence of material contact and clamping states. Based on the stroke zero-point anchor point or the stroke range endpoint anchor point, it performs translational or scaling correction on the preliminary estimation sequence of the gate stroke position. Combined with the contact section position constraint set and the clamping section position constraint set, it performs constraint projection and piecewise linear interpolation to form a constraint-consistent position sequence and outputs a calibration parameter update record. This enables online self-calibration and traceable updates of the gate stroke position real-time monitoring results, improving the reliability and verifiability of well control interlocking control in determining the true position of the gate. Attached Figure Description

[0069] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0070] Figure 1 This is a flowchart of a method for real-time monitoring and calibration of the internal gate position of a blowout preventer, as proposed in this invention.

[0071] Figure 2 This is a schematic diagram illustrating the identification of the true endpoint of a real-time monitoring and calibration method for the internal gate travel position of a blowout preventer proposed in this invention. Detailed Implementation

[0072] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, illustrating only the basic structure of the invention, and therefore only show the components relevant to the invention.

[0073] refer to Figures 1-2 A method for real-time monitoring and calibration of the travel position of the internal gate of a blowout preventer, comprising the following steps:

[0074] When the gate opening and closing action is triggered, the hydraulic pressure sequence, hydraulic flow sequence, valve control command sequence, action timing sequence and acoustic emission event stream are collected simultaneously;

[0075] The collected data is preprocessed to generate a set of multi-source standardized sequences of gate action.

[0076] Based on the multi-source standardized sequence set of gate action, a pressure-flow-command response trajectory is constructed and divided into the starting segment, constant speed segment, contact segment, clamping segment and arrival segment;

[0077] Extract segmented slope features, hysteresis morphology features, response delay features, and inter-segment transition features from the starting segment, constant speed segment, contact segment, pressing segment, and final position segment to generate a set of motion digital fingerprint vectors. Then, perform position inversion processing to obtain a preliminary estimation sequence of the gate travel position.

[0078] Within the candidate time window for the gate opening and closing action, identify the true arrival endpoint time and generate the zero point anchor point and the range endpoint anchor point of the stroke.

[0079] Perform event parameterization processing on the preprocessed acoustic emission event stream to identify the key moments of the gate contact section and the gate pressing section, and map them as the contact section position constraint set and the pressing section position constraint set;

[0080] Based on the zero-point anchor point of the travel range, the endpoint anchor point of the travel range, the set of position constraints of the contact section, the set of position constraints of the pressing section, and the preliminary estimation sequence of the gate travel position, the travel calibration process is performed to generate the real-time monitoring results of the gate travel position, the judgment results of the true position of the gate, and the update record of the calibration parameters.

[0081] In this embodiment, the hydraulic pressure sequence is a pressure data sequence formed by arranging the pressure sampling values ​​of the gate hydraulic execution circuit according to the sampling time during the gate opening and closing operation; the hydraulic flow sequence is a flow data sequence formed by arranging the flow sampling values ​​of the gate hydraulic execution circuit according to the sampling time during the gate opening and closing operation; the valve control command sequence is a command data sequence formed by arranging the valve control opening command, valve control reversing command, and valve control holding command output by the control system according to the command issuance time during the gate opening and closing operation; the action timing sequence is a timing event sequence representing the gate action start time, valve control command issuance time, valve group state switching time, and gate action end time during the gate opening and closing operation; and the acoustic emission event stream is an event data stream formed by arranging the acoustic emission events obtained by collecting acoustic emission signals and triggering detection during the gate opening and closing operation according to the event occurrence time. The acoustic emission event includes the acoustic emission event occurrence time timestamp, acoustic emission event energy, and acoustic emission event count rate.

[0082] In this embodiment, the generation of the multi-source standardized sequence set of gate action includes:

[0083] The start and end times of the gate action are determined based on the action timing sequence. The time window from the start to the end of the gate action is used as the gate action time window. Data segments corresponding to the hydraulic pressure sequence, hydraulic flow sequence, valve control command sequence and acoustic emission event stream are extracted within the gate action time window.

[0084] Using the start time of the gate action as a unified time reference, the time axes of the hydraulic pressure sequence, hydraulic flow sequence and valve control command sequence are uniformly shifted, and the timestamps of the occurrence time of acoustic emission events in the acoustic emission event stream are uniformly shifted to complete the alignment of the multi-source time axes.

[0085] Anomaly removal and pulsation suppression are performed on the hydraulic pressure sequence and hydraulic flow sequence, retaining the pressure change trend and flow change trend consistent with the gate opening and closing action response. Command jitter merging and holding interval merging are performed on the valve control command sequence to form a valve control command change trajectory consistent with the gate action time window. The valve control command change trajectory is the time sequence trajectory of the valve control command state changing with time, formed by arranging the valve control command sequence according to the command issuance time during the gate opening and closing action. The valve control command state includes valve control opening command, valve control reversing command, and valve control holding command.

[0086] Delete acoustic emission events outside the gate action time window and organize the time sequence to obtain the acoustic emission event stream within the gate action time window;

[0087] The processed hydraulic pressure sequence, hydraulic flow sequence, and valve control command change trajectory are merged to generate a set of multi-source standardized sequences of gate action.

[0088] In this embodiment, the division into the starting section, constant speed section, contact section, pressing section, and final position section specifically includes:

[0089] Obtain a multi-source standardized sequence set of gate action to obtain the hydraulic pressure sequence, hydraulic flow sequence, and valve control command change trajectory within the gate action time window;

[0090] According to the sampling time sequence, the hydraulic pressure sampling value, hydraulic flow sampling value and valve control command change trajectory corresponding to the command state at the same sampling time are combined into a triplet sequence to generate the pressure-flow-command response trajectory.

[0091] The pressure-flow-command response trajectory is a time-series response trajectory formed by combining the hydraulic pressure sequence sampling value, hydraulic flow sequence sampling value and valve control command state corresponding to the valve control command change trajectory at the same moment during the gate opening and closing operation.

[0092] Locate the inflection point when the pressure enters the continuously rising range from the initial stable range in the pressure-flow-command response trajectory, determine the end point of the starting segment, and mark the trajectory segment from the start of the gate action to the end point of the starting segment as the starting segment;

[0093] The initial stable interval is the pressure stable interval in which the hydraulic pressure sequence keeps the change amplitude not exceeding the preset threshold within a preset short time window after the start of the gate action. The continuous rising interval is the pressure rising interval in which the hydraulic pressure sequence shows a monotonically increasing increase at multiple consecutive sampling points and the cumulative increment exceeds the preset threshold.

[0094] After the end of the starting section, locate the trajectory segment where the hydraulic flow enters the stable platform range and the valve control command changes in the same control state. Mark this trajectory segment as the uniform speed segment, and determine the turning point when the hydraulic flow leaves the stable platform range as the end of the uniform speed segment.

[0095] The stable platform interval is the flow stability interval in which the hydraulic flow sequence maintains a change amplitude not exceeding a preset threshold at multiple consecutive sampling points. The valve control command change trajectory maintaining the same control state is the command holding interval in which the command status identifier of the valve control command change trajectory does not change within the corresponding time interval.

[0096] After the end of the uniform speed segment, locate the moment when the hydraulic pressure suddenly increases and the hydraulic flow simultaneously decreases, mark the trajectory segment from the end of the uniform speed segment to the moment of trajectory turning as the contact segment, and determine the moment of trajectory turning as the end of the contact segment.

[0097] The sudden increase in hydraulic pressure and the simultaneous decrease in hydraulic flow rate are the combined turning point characteristics where the pressure increment of the hydraulic pressure sequence exceeds a preset pressure change threshold and the flow rate decrease of the hydraulic flow rate sequence exceeds a preset flow rate decrease threshold within the same time window.

[0098] After the end of the contact section, locate the starting moment when the hydraulic pressure enters the high-level slow change range and the valve control command change trajectory enters the holding state. Mark the trajectory segment from the end of the contact section to the starting moment as the clamping section, and determine the starting moment as the end of the clamping section.

[0099] The high-level gradual change interval is the pressure gradual change interval in which the change amplitude does not exceed the preset gradual change threshold at multiple consecutive sampling points after the hydraulic pressure sequence reaches the preset high pressure threshold. The valve control command change trajectory entering the holding state is the command status identifier of the valve control command change trajectory switching to the valve control holding command and remaining unchanged in the corresponding time interval.

[0100] The trajectory segment from the end of the pressing section to the end of the gate action is marked as the positioning section, and the segment boundary times of the starting section, constant speed section, contact section, pressing section and positioning section are written into the pressure-flow-command response trajectory.

[0101] In this embodiment, the generation of the preliminary estimation sequence of the gate travel position specifically includes:

[0102] Extract the trajectory segments corresponding to the starting segment, constant speed segment, contact segment, clamping segment and arrival segment, and obtain the hydraulic pressure sequence, hydraulic flow sequence and valve control command change trajectory within each trajectory segment;

[0103] The slope characteristics of the starting segment are calculated for the trajectory segment of the starting segment, and the response delay characteristics of the starting segment are also calculated. The slope characteristics of the starting segment include the rising slope of the hydraulic pressure sequence and the changing slope of the hydraulic flow sequence. The response delay characteristics of the starting segment are the time difference between the command change time of the valve control command change trajectory and the time difference between the hydraulic pressure sequence entering the continuous rising interval.

[0104] The slope characteristics and hysteresis morphology characteristics of the uniform speed segment trajectory are calculated, and the response delay characteristics of the uniform speed segment are also calculated. The slope characteristics of the uniform speed segment are the slopes of pressure change and flow change of hydraulic pressure sequence and hydraulic flow sequence over time within the uniform speed segment. The hysteresis morphology characteristics of the uniform speed segment are composed of the area of ​​pressure-flow hysteresis loop formed by hydraulic pressure sequence and hydraulic flow sequence within the uniform speed segment and the hysteresis loop offset. The response delay characteristics of the uniform speed segment are the time difference between the command change time of valve control command change trajectory and the time difference between the hydraulic flow sequence entering the stable plateau interval.

[0105] The pressure-flow hysteresis loop area is the area value calculated by performing area calculation on the region enclosed by the closed hysteresis loop after constructing a pressure-flow plane trajectory with the hydraulic flow sequence as the horizontal axis and the hydraulic pressure sequence as the vertical axis within the uniform speed segment. The hysteresis loop offset is the offset of the geometric center of the closed hysteresis loop relative to the preset reference point in the pressure axis direction and the flow axis direction. The closed hysteresis loop is a closed curve formed by connecting the beginning and end of the round-trip trajectories formed by the hydraulic pressure sequence and the hydraulic flow sequence in the pressure-flow plane within the uniform speed segment. The preset reference point is the pressure-flow plane reference coordinate point composed of the hydraulic pressure sampling value and the hydraulic flow sampling value corresponding to the beginning time of the uniform speed segment.

[0106] The contact segment trajectory segment is used to calculate the contact segment slope characteristics and inter-segment transition characteristics, and the contact segment response delay characteristics are also calculated. The contact segment slope characteristics are the slope of the pressure change of the hydraulic pressure sequence and the slope of the flow change of the hydraulic flow sequence within the contact segment. The inter-segment transition characteristics include the transition duration and transition amplitude at the moment corresponding to the joint turning point where the hydraulic pressure suddenly increases and the hydraulic flow simultaneously decreases from the end of the uniform speed segment. The transition duration is the time difference between the end of the uniform speed segment and the moment corresponding to the joint turning point where the hydraulic pressure suddenly increases and the hydraulic flow simultaneously decreases. The transition amplitude is the pressure increment of the hydraulic pressure sequence and the flow decrease of the hydraulic flow sequence within the time difference. The contact segment response delay characteristics are the time difference between the moment corresponding to the joint turning point and the moment of the most recent command change of the valve control command trajectory.

[0107] The hysteresis morphology and inter-segment transition characteristics of the clamping segment are calculated for the clamping segment trajectory segment, and the response delay characteristics of the clamping segment are calculated. The inter-segment transition characteristics of the clamping segment include the transition time and pressure increment from the end of the contact segment to the hydraulic pressure sequence entering the high-level slow change interval and the valve control command change trajectory entering the holding state. The response delay characteristics of the clamping segment are the time difference between the moment when the valve control command change trajectory enters the holding state and the moment when the hydraulic pressure sequence enters the high-level slow change interval.

[0108] The hysteresis morphology of the pressing section is characterized by constructing a pressure-flow plane trajectory within the pressing section trajectory segment with the hydraulic flow sequence as the horizontal axis and the hydraulic pressure sequence as the vertical axis, forming a closed hysteresis loop, and then calculating the area of ​​the region enclosed by the closed hysteresis loop. The hysteresis loop offset is obtained by determining the geometric center of the closed hysteresis loop with the mean pressure coordinates and the mean flow coordinates corresponding to the closed hysteresis loop, and calculating the offset of the geometric center of the closed hysteresis loop relative to a preset reference point. The preset reference point is the pressure-flow plane reference coordinate point composed of the hydraulic pressure sampling value and the hydraulic flow sampling value corresponding to the start time of the pressing section.

[0109] The slope characteristics of the arrival segment and the transition characteristics between the arrival segments are calculated for the trajectory segments of the arrival segment. The slope characteristics of the arrival segment are the slope of pressure change and the slope of flow change of hydraulic pressure sequence and hydraulic flow sequence within the time range of the arrival segment. The transition characteristics between the arrival segments include the transition time from the end of the pressing segment to the end of the gate action and the pressure gradual change amplitude value. The pressure gradual change amplitude value is the difference between the maximum pressure sampling value and the minimum pressure sampling value of the hydraulic pressure sequence within the time range of the arrival segment.

[0110] According to the sequence of starting segment, constant speed segment, contact segment, pressing segment and arrival segment, the segment slope characteristics, hysteresis morphology characteristics, response delay characteristics and inter-segment transition characteristics are spliced ​​together to form a set of motion digital fingerprint vectors.

[0111] Each gate opening and closing action is divided into five segments on the pressure-flow-command response trajectory: starting segment, constant speed segment, contact segment, pressing segment, and arrival segment. Segment slope features, hysteresis morphology features, response delay features, and inter-segment transition features are extracted from the five segments and spliced ​​together to form a motion digital fingerprint vector. The motion digital fingerprint vector is repeatedly generated for multiple gate opening and closing actions to obtain multiple motion digital fingerprint vectors and form a set of motion digital fingerprint vectors.

[0112] Position inversion is performed based on the set of motion digital fingerprint vectors to generate a preliminary estimation sequence of the gate travel position.

[0113] In this embodiment, the execution location inversion specifically includes:

[0114] The motion digital fingerprint vector set is extracted according to the starting segment, constant speed segment, contact segment, pressing segment and arrival segment to obtain five feature groups, and then normalized to obtain normalized five feature groups.

[0115] Perform feature extraction and normalization on each historical motion digital fingerprint vector in the motion digital fingerprint database according to the starting segment, constant speed segment, contact segment, pressing segment and positioning segment to obtain a historical normalized five-segment feature group.

[0116] The motion digital fingerprint database is generated by preprocessing hydraulic pressure sequences, hydraulic flow sequences, valve control command sequences, action timing sequences, and acoustic emission event streams obtained from multiple gate opening and closing actions to create a set of multi-source standardized sequence of gate actions. Based on the set of multi-source standardized sequence of gate actions, a pressure-flow-command response trajectory is constructed and segmented into starting segment, constant speed segment, contact segment, pressing segment, and arrival segment. Segment slope features, hysteresis morphology features, response delay features, and inter-segment transition features are extracted and spliced ​​to form a set of historical motion digital fingerprint vectors. Finally, a historical data set is formed by binding each historical motion digital fingerprint vector to a corresponding gate stroke position reference sequence and summarizing them.

[0117] For each segment, calculate the feature difference and perform a weighted sum to obtain the segment consistency score. Sum the five segment consistency scores to obtain the total segment consistency score set.

[0118] Select the target historical motion digital fingerprint vector corresponding to the minimum segment consistency total score, and read the gate travel position reference sequence corresponding to the target historical motion digital fingerprint vector;

[0119] The gate travel position reference sequence is obtained during the motion digital fingerprint database establishment phase. During the motion digital fingerprint database establishment phase, the gate opening and closing action is performed under maintenance or debugging conditions. Simultaneously, hydraulic pressure sequence, hydraulic flow sequence, valve control command sequence, action timing sequence, and acoustic emission event stream are collected and a motion digital fingerprint vector set is generated. At the same time, the actual value of the gate travel position is recorded according to the sampling time based on the gate travel gauge reading and arranged in chronological order to form a gate travel position reference sequence. The motion digital fingerprint vector set and the gate travel position reference sequence are associated one by one and written into the motion digital fingerprint database.

[0120] The gate travel position reference sequence is resampled by linear interpolation according to the sampling time of the gate action time window to generate a preliminary estimation sequence of the gate travel position.

[0121] In this embodiment, the generation of the zero-point anchor point and the end-point anchor point of the travel range specifically includes:

[0122] Obtain the multi-source standardized sequence set of gate action and the pressure-flow-command response trajectory with completed segment marking, read the start time of the arrival segment and the end time of the gate action, and use the start time of the arrival segment to the end time of the gate action to determine the candidate time window for arrival;

[0123] Within the candidate time window, the pressure waveform segment corresponding to the hydraulic pressure sequence is extracted and high-frequency extraction processing is performed to obtain the high-frequency response waveform of hydraulic pressure. The flow waveform segment corresponding to the hydraulic flow sequence is extracted and transient extraction processing is performed to obtain the transient response waveform of hydraulic flow.

[0124] The high-frequency extraction process uses a bandpass filter, and the transient extraction process uses first-order difference processing.

[0125] For each sampling point of the high-frequency response waveform of hydraulic pressure, the adjacent difference is calculated to generate a sorted sequence of sampling point pairs. The sampling point pair with the largest absolute difference is selected, and the next sampling point of the sampling point pair is defined as the pressure change moment. The pressure change amplitude is generated, which is the absolute value of the difference between the hydraulic pressure sampling value corresponding to the pressure change moment and the hydraulic pressure sampling value corresponding to the previous sampling moment. The values ​​of the sampling point pair difference sorted sequence are sorted from high to low.

[0126] A ringing analysis window of a preset length is extracted with the pressure change moment as the center. The set of adjacent peak moments of the high-frequency response waveform of hydraulic pressure within the ringing analysis window is located, and the ringing frequency feature is calculated. The ringing frequency feature is the mean and reciprocal of the difference between adjacent peak moments. The set of adjacent peak moments is the set of peak moments corresponding to multiple peaks detected in the high-frequency response waveform of hydraulic pressure within the ringing analysis window in chronological order. It is used to calculate the difference between adjacent peak moments and generate the ringing frequency feature.

[0127] Within the ringing analysis window, the ratio of the first peak amplitude to the last peak amplitude is calculated to obtain the attenuation ratio, and the attenuation duration is generated to obtain the ringing attenuation characteristics, which include the attenuation ratio and the attenuation duration.

[0128] The difference between the minimum value of the transient response waveform of hydraulic flow and the steady-state baseline value is calculated in the neighborhood of the moment of pressure change, and the flow rate drop amplitude is generated.

[0129] The steady-state baseline value is the average flow sampling value of the hydraulic flow transient response waveform within the candidate time window before the flow decline occurs. The flow decline is the transient decline behavior in which the hydraulic flow transient response waveform within the candidate time window shows a significant decrease relative to the steady-state baseline value and the flow reduction exceeds the preset flow decline threshold.

[0130] Within the neighborhood of the pressure change moment, extract the pressure change amplitude, ringing frequency features, ringing attenuation features, and flow rate decline amplitude, and normalize them to construct the arrival transient feature vector. Calculate the vector distance between the arrival transient feature vector and the preset true arrival transient template vector. If the vector distance exceeds the preset distance threshold, replace the pressure change moment with the next sampling point moment corresponding to the next index in the sampling point difference sorting sequence and repeat the above steps until the vector distance does not exceed the preset distance threshold.

[0131] The preset distance threshold is calculated from the set of vector distance values ​​between the true arrival endpoint time sample collected during the motion digital fingerprint database establishment phase and the preset true arrival transient template vector. The mean and standard deviation of the vector distance value set are taken to generate the threshold and written into the threshold configuration item. The preset true arrival transient template vector is the standard arrival transient feature vector obtained by averaging the corresponding feature components of the arrival transient feature vector corresponding to the true arrival endpoint time sample collected during the motion digital fingerprint database establishment phase.

[0132] The moment of pressure change that satisfies the vector distance value not exceeding the preset distance threshold is determined as the true arrival endpoint moment. The gate opening direction or gate closing direction is determined according to the command status identifier of the valve control command change trajectory. The true arrival endpoint moment corresponding to the gate closing direction is mapped as the stroke zero point anchor point, and the true arrival endpoint moment corresponding to the gate opening direction is mapped as the stroke range endpoint anchor point. The stroke zero point anchor point is the true arrival endpoint moment corresponding to the gate closing direction, and the stroke range endpoint anchor point is the true arrival endpoint moment corresponding to the gate opening direction.

[0133] In this embodiment, the generation of the contact segment position constraint set and the clamping segment position constraint set specifically includes:

[0134] The acoustic emission event stream is extracted from the set of multi-source standardized sequences of gate action, and the acoustic emission event stream is sorted according to the timestamp of the occurrence time of the acoustic emission event to form a time-ordered acoustic emission event sequence.

[0135] Perform event parameterization processing on the time-ordered acoustic emission event sequence, representing each acoustic emission event as an acoustic emission event parameter consisting of the timestamp of the occurrence time of the acoustic emission event, the energy of the acoustic emission event, and the count rate of the acoustic emission event, and sum them up to form an acoustic emission event parameter set;

[0136] Statistical analysis is performed on the set of acoustic emission event parameters to generate an acoustic emission event count rate sequence and an acoustic emission event energy sequence. The acoustic emission event count rate sequence and the acoustic emission event energy sequence are then aligned by timestamps to form a joint acoustic emission response sequence.

[0137] Within the contact time range of the pressure-flow-command response trajectory, calculate the energy increment and count rate increment of adjacent time windows for the acoustic emission joint response sequence, locate the timestamp of the acoustic emission event when the energy increment and count rate increment simultaneously reach local peaks, and generate a set of key moments of the gate contact segment.

[0138] Within the time range of the pressure-flow-command response trajectory, the overlapping period of the energy accumulation rise interval and the count rate sustained high interval is calculated for the acoustic emission joint response sequence. The timestamp of the acoustic emission event corresponding to the energy increment reaching the local peak within the overlapping period is located, and a set of key moments of the gate clamping section is generated.

[0139] The sustained high count rate interval is the time interval in which the count rate sequence of acoustic emission events is not lower than a preset high count rate threshold at multiple consecutive sampling times within the compression period.

[0140] The key moment set of the gate contact section and the key moment set of the gate pressing section are merged to form a contact anchor point set. The contact anchor point set is then mapped to the corresponding sampling position of the gate stroke position preliminary estimation sequence according to the sampling time of the gate action time window, generating a contact section position constraint set and a pressing section position constraint set. The contact section position constraint set is the set of contact section position constraint values ​​obtained by mapping the contact anchor point set to the gate stroke position preliminary estimation sequence, and the pressing section position constraint set is the set of pressing section position constraint values ​​obtained by mapping the contact anchor set to the gate stroke position preliminary estimation sequence.

[0141] In this embodiment, the execution stroke calibration process specifically includes:

[0142] Obtain the preliminary estimation sequence of the gate travel position, the zero point anchor point or the end point anchor point of the travel range, the contact section position constraint set and the clamping section position constraint set, and expand the preliminary estimation sequence of the gate travel position according to the sampling time to form a position time series array;

[0143] If a zero-point anchor point exists, the position value corresponding to the sampling time of the zero-point anchor point is read, the difference between the position value and the zero-point position value is calculated as the translation correction amount, and the translation correction amount is synchronously applied to all position values ​​in the position time sequence array to obtain the anchor point aligned position time sequence array;

[0144] If there is a travel range endpoint anchor point, read the position value corresponding to the sampling time of the travel range endpoint anchor point, calculate the ratio of the range endpoint position value to the position value as the scaling correction amount, and apply the scaling correction amount synchronously to all position values ​​in the position time sequence array to obtain the anchor point aligned position time sequence array.

[0145] Based on the anchor point alignment position time sequence array, constraint projection processing is performed on the contact section position constraint set and the clamping section position constraint set. The position value corresponding to the sampling time of each constraint value is replaced with the constraint position value. Piecewise linear interpolation is performed in the preset neighborhood before and after the constraint sampling time to generate a constraint consistent position sequence.

[0146] The constrained consistent position sequence is used as the real-time monitoring result of the gate travel position. Based on the travel zero point anchor point or travel range endpoint anchor point corresponding to the true arrival endpoint, the gate true arrival state judgment result is generated. At the same time, the translation correction amount or scaling correction amount is recorded, and the calibration parameter update record is generated.

[0147] In this embodiment, the zero point position value is the zero point calibration value of the gate travel position, and the range endpoint position value is the maximum travel calibration value of the gate travel position.

[0148] Example 1: To verify the feasibility and applicability of the present invention in actual working conditions, the present invention was applied to the operation of a blowout preventer (BOP) system on an offshore drilling platform. This BOP operates under high pressure, high load, and frequent opening and closing conditions. During repeated shut-off and opening processes, the gate is susceptible to hydraulic pulsations, structural vibrations, and changes in ambient temperature. This makes it difficult for traditional stroke monitoring methods relying on displacement sensors or limit signals to continuously and accurately reflect the true position of the gate. Especially when the gate contacts the wellhead sealing surface and enters the clamping stage, there are problems such as misjudgment of the true position and the inability to correct stroke parameters online due to time drift, thus affecting the reliability of well control interlocking.

[0149] In this application scenario, when the well control system issues a gate opening and closing command, the method of this invention simultaneously collects the hydraulic pressure sequence, hydraulic flow sequence, valve control command sequence, action timing sequence, and acoustic emission event stream in the gate's hydraulic execution circuit. Under a unified time reference, the multi-source data is preprocessed to form a standardized multi-source sequence set of gate actions. Based on this set, a pressure-flow-command response trajectory is constructed, and the entire gate opening and closing process is automatically divided into a starting segment, a constant speed segment, a contact segment, a clamping segment, and a positioning segment. This allows the gate's response state from the start of action to its final positioning to be continuously characterized. Within each segment, segment slope features, hysteresis morphology features, response delay features, and inter-segment transition features are extracted to form a motion digital fingerprint vector set. A preliminary estimation sequence of the gate's stroke position is generated through position inversion, thereby obtaining a stable stroke change trend without direct displacement measurement.

[0150] During the stage when the gate is about to complete its opening and closing action, this invention performs joint analysis on the high-frequency response of hydraulic pressure and the transient response of hydraulic flow within the candidate time window for arrival. It constructs a transient feature vector for arrival by using transient features such as pressure change, ringing response, and flow drop, and matches it with a preset true arrival transient template to accurately identify the true arrival endpoint of the gate, thereby generating the zero-point anchor point or the end point anchor point of the travel range. At the same time, by parameterizing the acoustic emission event flow, it identifies the key moments corresponding to the contact and clamping behavior of the gate material in the contact and clamping sections, and maps them to the position constraint set of the contact section and the position constraint set of the clamping section. This makes the travel calibration process not only dependent on hydraulic response information, but also introduces independent constraint evidence at the structural contact level.

[0151] In actual operation, the zero point anchor point or the end point anchor point of the stroke range, together with the contact section position constraint set and the clamping section position constraint set, are applied to the preliminary estimation sequence of the gate stroke position. The stroke calibration process is then performed on it to obtain the real-time monitoring result of the gate stroke position, and the gate true position determination result and calibration parameter update record are output simultaneously.

[0152] To verify the performance of the present invention in practice, it was compared with traditional methods, and the results are shown in Table 1.

[0153] Table 1. Overall Performance Comparison Table

[0154]

[0155] From the perspective of stroke position estimation accuracy, traditional methods based on limit switches or valve position signals have an average stroke error of 0.35 mm and a maximum stroke drift of 0.66 mm. This indicates that such methods are highly sensitive to mechanical clearance, installation deviation, and long-term operational drift, making it difficult to reflect the true movement state of the gate. In contrast, the method of this invention controls the average stroke error to 0.09 mm and the maximum drift to only 0.14 mm, which is a significant reduction. This shows that by constructing a stroke inversion model using multi-source hydraulic response characteristics and motion digital fingerprints, the systematic error accumulation caused by a single signal can be effectively suppressed.

[0156] Regarding the accuracy of true position determination, while the traditional single-threshold pressure method outperforms simple limit switch determination, its accuracy remains at 91.2%, indicating a significant margin for error in scenarios involving hydraulic fluctuations or valve control command jitter. The method of this invention achieves a true position determination accuracy of 95.7%, with a false judgment rate of only 3.2% under abnormal operating conditions, lower than traditional methods. This difference directly reflects the advantages of the multi-dimensional transient position determination mechanism, which, through the coordinated constraints of pressure surges, flow rate drops, and acoustic emission key events, makes position identification no longer dependent on a single physical quantity, thereby improving the reliability of the determination.

[0157] From the perspective of time stability, the traditional method has a time fluctuation range of 45 ms and 36 ms, respectively, indicating that the judgment results have obvious time instability problems under different working conditions. However, the judgment time fluctuation range of the method of this invention is reduced to 16 ms, which shows that by limiting the candidate time window for arrival and focusing the judgment on transient features, the dispersion of the judgment time can be effectively compressed, making the identification of the true arrival time more concentrated and stable, which is beneficial to subsequent travel anchor point mapping and calibration processing.

[0158] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for real-time monitoring and calibration of the stroke position of the internal gate of a blowout preventer, characterized in that, Includes the following steps: When the gate opening and closing action is triggered, the hydraulic pressure sequence, hydraulic flow sequence, valve control command sequence, action timing sequence and acoustic emission event stream are collected simultaneously; The collected data is preprocessed to generate a set of multi-source standardized sequences of gate action. Based on the multi-source standardized sequence set of gate action, a pressure-flow-command response trajectory is constructed and divided into the starting segment, constant speed segment, contact segment, clamping segment and arrival segment; Extract segmented slope features, hysteresis morphology features, response delay features, and inter-segment transition features from the starting segment, constant speed segment, contact segment, pressing segment, and final position segment to generate a set of motion digital fingerprint vectors. Then, perform position inversion processing to obtain a preliminary estimation sequence of the gate travel position. Within the candidate time window for the gate opening and closing action, identify the true arrival endpoint time and generate the zero point anchor point and the range endpoint anchor point of the stroke. Perform event parameterization processing on the preprocessed acoustic emission event stream to identify the key moments of the gate contact section and the gate pressing section, and map them as the contact section position constraint set and the pressing section position constraint set; Based on the zero-point anchor point of the stroke, the end point anchor point of the stroke range, the set of position constraints of the contact section, the set of position constraints of the pressing section, and the preliminary estimation sequence of the gate stroke position, the stroke calibration process is performed to generate the real-time monitoring result of the gate stroke position, the judgment result of the true position of the gate, and the update record of the calibration parameters. The generation of the zero-point anchor point and the end-point anchor point of the travel range specifically includes: Obtain the multi-source standardized sequence set of gate action and the pressure-flow-command response trajectory with completed segment marking, read the start time of the arrival segment and the end time of the gate action, and use the start time of the arrival segment to the end time of the gate action to determine the candidate time window for arrival; Within the candidate time window, the pressure waveform segment corresponding to the hydraulic pressure sequence is extracted and high-frequency extraction processing is performed to obtain the high-frequency response waveform of hydraulic pressure. The flow waveform segment corresponding to the hydraulic flow sequence is extracted and transient extraction processing is performed to obtain the transient response waveform of hydraulic flow. For each sampling point of the high-frequency response waveform of hydraulic pressure, the adjacent difference is calculated, a sorted sequence of sampling point pairs is generated, and the sampling point pair with the largest absolute value of the difference is selected. The time of the next sampling point of the sampling point pair is defined as the pressure change time, and the pressure change amplitude is generated. A ringing analysis window of a preset length is extracted with the moment of pressure change as the center. The set of adjacent peak moments of the high-frequency response waveform of hydraulic pressure within the ringing analysis window is located, and the ringing frequency characteristics are calculated. Within the ringing analysis window, the ratio of the first peak amplitude to the last peak amplitude is calculated to obtain the attenuation ratio, and the attenuation duration is generated to obtain the ringing attenuation characteristics. The difference between the minimum value of the transient response waveform of hydraulic flow and the steady-state baseline value is calculated in the neighborhood of the moment of pressure change, and the flow rate drop amplitude is generated. Within the neighborhood of the pressure change moment, extract the pressure change amplitude, ringing frequency features, ringing attenuation features, and flow rate decline amplitude, and normalize them to construct the arrival transient feature vector. Calculate the vector distance between the arrival transient feature vector and the preset true arrival transient template vector. If the vector distance exceeds the preset distance threshold, replace the pressure change moment with the next sampling point moment corresponding to the next index in the sampling point difference sorting sequence and repeat the above steps until the vector distance does not exceed the preset distance threshold. The moment of pressure change that satisfies the vector distance value not exceeding the preset distance threshold is determined as the true end point moment. The gate opening direction or gate closing direction is determined according to the command status identifier of the valve control command change trajectory. The true end point moment corresponding to the gate closing direction is mapped as the stroke zero point anchor point. The true end point moment corresponding to the gate opening direction is mapped as the stroke range end point anchor point.

2. The method for real-time monitoring and calibration of the internal gate travel position of a blowout preventer according to claim 1, characterized in that, The hydraulic pressure sequence is a pressure data sequence formed by arranging the pressure sampling values ​​of the gate hydraulic execution circuit according to the sampling time during the gate opening and closing operation. The hydraulic flow sequence is a flow data sequence formed by arranging the flow sampling values ​​of the gate hydraulic execution circuit according to the sampling time during the gate opening and closing operation. The valve control command sequence is a command data sequence formed by arranging the valve control opening command, valve control reversing command, and valve control holding command output by the control system according to the command issuance time during the gate opening and closing operation. The action timing sequence is a timing event sequence representing the gate action start time, valve control command issuance time, valve group state switching time, and gate action end time during the gate opening and closing operation. The acoustic emission event stream is an event data stream formed by arranging the acoustic emission events obtained by collecting acoustic emission signals and triggering detection during the gate opening and closing operation according to the event occurrence time. The acoustic emission event includes the acoustic emission event occurrence time timestamp, acoustic emission event energy, and acoustic emission event count rate.

3. The method for real-time monitoring and calibration of the internal gate travel position of a blowout preventer according to claim 1, characterized in that, The generation of the multi-source standardized sequence set of gate action includes: The start and end times of the gate action are determined based on the action timing sequence. The time window from the start to the end of the gate action is used as the gate action time window. Data segments corresponding to the hydraulic pressure sequence, hydraulic flow sequence, valve control command sequence and acoustic emission event stream are extracted within the gate action time window. Using the start time of the gate action as a unified time reference, the time axes of the hydraulic pressure sequence, hydraulic flow sequence and valve control command sequence are uniformly shifted, and the timestamps of the occurrence time of acoustic emission events in the acoustic emission event stream are uniformly shifted to complete the alignment of the multi-source time axes. Anomaly removal and pulsation suppression are performed on the hydraulic pressure sequence and hydraulic flow sequence, and command jitter merging and holding interval merging are performed on the valve control command sequence to form a valve control command change trajectory consistent with the gate action time window; Delete acoustic emission events outside the gate action time window and organize the time sequence to obtain the acoustic emission event stream within the gate action time window; The processed hydraulic pressure sequence, hydraulic flow sequence, and valve control command change trajectory are merged to generate a set of multi-source standardized sequences of gate action.

4. The method for real-time monitoring and calibration of the stroke position of the internal gate of a blowout preventer according to claim 1, characterized in that, The division into the starting section, constant speed section, contact section, pressing section, and final position section specifically includes: Obtain a multi-source standardized sequence set of gate action to obtain the hydraulic pressure sequence, hydraulic flow sequence, and valve control command change trajectory within the gate action time window; According to the sampling time sequence, the hydraulic pressure sampling value, hydraulic flow sampling value and valve control command change trajectory corresponding to the command state at the same sampling time are combined into a triplet sequence to generate the pressure-flow-command response trajectory. Locate the inflection point when the pressure enters the continuously rising range from the initial stable range in the pressure-flow-command response trajectory, determine the end point of the starting segment, and mark the trajectory segment from the start of the gate action to the end point of the starting segment as the starting segment; After the end of the starting section, locate the trajectory segment where the hydraulic flow enters the stable platform range and the valve control command changes in the same control state. Mark this trajectory segment as the uniform speed segment, and determine the turning point when the hydraulic flow leaves the stable platform range as the end of the uniform speed segment. After the end of the uniform speed segment, locate the moment when the hydraulic pressure suddenly increases and the hydraulic flow simultaneously decreases, mark the trajectory segment from the end of the uniform speed segment to the moment of trajectory turning as the contact segment, and determine the moment of trajectory turning as the end of the contact segment. After the end of the contact section, locate the starting moment when the hydraulic pressure enters the high-level slow change range and the valve control command change trajectory enters the holding state. Mark the trajectory segment from the end of the contact section to the starting moment as the clamping section, and determine the starting moment as the end of the clamping section. The trajectory segment from the end of the pressing section to the end of the gate action is marked as the positioning section, and the segment boundary times of the starting section, constant speed section, contact section, pressing section and positioning section are written into the pressure-flow-command response trajectory.

5. The method for real-time monitoring and calibration of the stroke position of the internal gate of a blowout preventer according to claim 1, characterized in that, The generation of the preliminary estimation sequence of the gate travel position specifically includes: Extract the trajectory segments corresponding to the starting segment, constant speed segment, contact segment, clamping segment and arrival segment, and obtain the hydraulic pressure sequence, hydraulic flow sequence and valve control command change trajectory within each trajectory segment; Calculate the slope characteristics of the starting segment for the trajectory segment and the response delay characteristics of the starting segment; Calculate the segmented slope characteristics and hysteresis morphology characteristics of the uniform velocity segment for the trajectory segment of the uniform velocity segment, and calculate the response delay characteristics of the uniform velocity segment. Calculate the contact segment slope characteristics and inter-segment transition characteristics for the contact segment trajectory segment, and calculate the contact segment response delay characteristics; The hysteresis morphology and inter-segment transition characteristics of the compaction segment are calculated for the trajectory segment of the compaction segment, and the response delay characteristics of the compaction segment are also calculated. Calculate the segment slope characteristics and inter-segment transition characteristics of the trajectory segments in the arrival segment; According to the sequence of starting segment, constant speed segment, contact segment, pressing segment and arrival segment, the segment slope characteristics, hysteresis morphology characteristics, response delay characteristics and inter-segment transition characteristics are spliced ​​together to form a set of motion digital fingerprint vectors. Position inversion is performed based on the set of motion digital fingerprint vectors to generate a preliminary estimation sequence of the gate travel position.

6. The method for real-time monitoring and calibration of the stroke position of the internal gate of a blowout preventer according to claim 5, characterized in that, The execution location inversion specifically includes: The motion digital fingerprint vector set is extracted according to the starting segment, constant speed segment, contact segment, pressing segment and arrival segment to obtain five feature groups, and then normalized to obtain normalized five feature groups. Perform feature extraction and normalization on each historical motion digital fingerprint vector in the motion digital fingerprint database according to the starting segment, constant speed segment, contact segment, pressing segment and positioning segment to obtain a historical normalized five-segment feature group. For each segment, calculate the feature difference and perform a weighted sum to obtain the segment consistency score. Sum the five segment consistency scores to obtain the total segment consistency score set. Select the target historical motion digital fingerprint vector corresponding to the minimum segment consistency total score, and read the gate travel position reference sequence corresponding to the target historical motion digital fingerprint vector; The gate travel position reference sequence is resampled by linear interpolation according to the sampling time of the gate action time window to generate a preliminary estimation sequence of the gate travel position.

7. The method for real-time monitoring and calibration of the internal gate travel position of a blowout preventer according to claim 1, characterized in that, The generation of the contact segment position constraint set and the clamping segment position constraint set specifically includes: The acoustic emission event stream is extracted from the set of multi-source standardized sequences of gate action, and the acoustic emission event stream is sorted according to the timestamp of the occurrence time of the acoustic emission event to form a time-ordered acoustic emission event sequence. Perform event parameterization processing on the time-ordered acoustic emission event sequence, representing each acoustic emission event as an acoustic emission event parameter consisting of the timestamp of the occurrence time of the acoustic emission event, the energy of the acoustic emission event, and the count rate of the acoustic emission event, and sum them up to form an acoustic emission event parameter set; Statistical analysis is performed on the set of acoustic emission event parameters to generate an acoustic emission event count rate sequence and an acoustic emission event energy sequence. The acoustic emission event count rate sequence and the acoustic emission event energy sequence are then aligned by timestamps to form a joint acoustic emission response sequence. Within the contact time range of the pressure-flow-command response trajectory, calculate the energy increment and count rate increment of adjacent time windows for the acoustic emission joint response sequence, locate the timestamp of the acoustic emission event when the energy increment and count rate increment simultaneously reach local peaks, and generate a set of key moments of the gate contact segment. Within the time range of the pressure-flow-command response trajectory, the overlapping period of the energy accumulation rise interval and the count rate sustained high interval is calculated for the acoustic emission joint response sequence. The timestamp of the acoustic emission event corresponding to the energy increment reaching the local peak within the overlapping period is located, and a set of key moments of the gate clamping section is generated. The key moment set of the gate contact section and the key moment set of the gate pressing section are merged to form a contact anchor point set. The contact anchor point set is then mapped to the corresponding sampling position of the preliminary estimation sequence of the gate stroke position according to the sampling time of the gate action time window, thereby generating a contact section position constraint set and a pressing section position constraint set.

8. The method for real-time monitoring and calibration of the internal gate travel position of a blowout preventer according to claim 1, characterized in that, The execution stroke calibration process specifically includes: Obtain the preliminary estimation sequence of the gate travel position, the zero point anchor point or the end point anchor point of the travel range, the contact section position constraint set and the clamping section position constraint set, and expand the preliminary estimation sequence of the gate travel position according to the sampling time to form a position time series array; If a zero-point anchor point exists, the position value corresponding to the sampling time of the zero-point anchor point is read, the difference between the position value and the zero-point position value is calculated as the translation correction amount, and the translation correction amount is synchronously applied to all position values ​​in the position time sequence array to obtain the anchor point aligned position time sequence array; If there is a travel range endpoint anchor point, read the position value corresponding to the sampling time of the travel range endpoint anchor point, calculate the ratio of the range endpoint position value to the position value as the scaling correction amount, and apply the scaling correction amount synchronously to all position values ​​in the position time sequence array to obtain the anchor point aligned position time sequence array. Based on the anchor point alignment position time sequence array, constraint projection processing is performed on the contact section position constraint set and the clamping section position constraint set. The position value corresponding to the sampling time of each constraint value is replaced with the constraint position value. Piecewise linear interpolation is performed in the preset neighborhood before and after the constraint sampling time to generate a constraint consistent position sequence. The constrained consistent position sequence is used as the real-time monitoring result of the gate travel position. Based on the travel zero point anchor point or travel range endpoint anchor point corresponding to the true arrival endpoint, the gate true arrival state judgment result is generated. At the same time, the translation correction amount or scaling correction amount is recorded, and the calibration parameter update record is generated.

9. A method for real-time monitoring and calibration of the stroke position of the internal gate of a blowout preventer according to claim 8, characterized in that, The zero-point position value is the zero-point calibration value of the gate travel position, and the range endpoint position value is the maximum travel calibration value of the gate travel position.

Images