Valve detection control method based on front end and rear end

Abstract

The invention discloses a valve detection control method based on a front end and a rear end, particularly relates to the field of control valve online detection, and is used for solving the problem that state degradation of a stock valve in long-term operation is difficult to quantify in time and acts on real-time control. Small-stroke test is executed at a front-end terminal, the angular displacement of a valve rod and the driving amount of an executing mechanism are synchronously collected under the unified time reference, and stroke segmentation and segment-level characteristic parameter sets are obtained through incremental segmentation fitting; the back-end platform performs modeling on the segment-level characteristic parameter set and the process constraint data in the stroke segment dimension to generate a health constraint parameter set and issues the health constraint parameter set to the front-end terminal, and the front-end terminal performs angle projection and action speed limitation on a target opening instruction according to the health constraint parameter set. And the integrated management of the online diagnosis result and the control behavior of the stock valve is realized by combining the angle deviation continuous monitoring, the rollback control based on the historical safety position and the abnormal event reporting.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of online detection of control valves, and more particularly, to a valve detection control method based on front and back ends. BACKGROUND

[0002] In long-term running pipe network systems such as water, gas, oil product transportation and distribution, and industrial devices, a large number of in-service valves are gate valves, ball valves, butterfly valves or stop valves configured on site according to working conditions many years ago. The nameplate information, structural parameters and factory characteristic data thereof are often incomplete, and the original settings of some devices have been difficult to verify after multiple overhauls and modifications. These inventory valves are often in a high-pressure or corrosive medium environment, and once they have slow action, jamming, poor sealing or misoperation, they may cause pressure imbalance, key unit shutdown, leakage expansion and increased energy consumption, which poses a hidden danger to production safety and public service stability. Under the background of rising labor costs and stricter supervision, the operation and maintenance unit hopes to configure online detection and remote control functions for inventory valves without stopping production or with as little stopping production as possible, so as to realize the visualization of the running state and the advance control of risks.

[0003] In the prior art, for newly built or parameter clear control valves, digital valve positioners, intelligent actuators and partial stroke test devices for safety shutdown valves have been widely used. Under the premise of known valve type and nominal stroke, the stroke endpoints, friction state and execution capability are obtained through full stroke self-calibration or pre-stroke test, and the diagnostic results are uploaded to the upper computer or asset management system for analysis. At the same time, non-invasive monitoring devices installed in the valve handle, valve stem or pipeline outer wall in a clamping or external pasting manner have also appeared, which are used to collect opening and closing states or vibration signals to realize remote state monitoring. These schemes work well on newly built or parameter clear devices, but in the online upgrading scene of large-scale, various types and missing design data inventory valves, if full stroke self-calibration or large stroke test is directly used, it is easy to cause pressure fluctuation and flow mutation in the pressure running pipe network, which is difficult to execute frequently in the field. If only the above non-invasive monitoring devices are relied on, the relationship between the driving amount and the valve stem displacement in different opening intervals is not described in detail, it is difficult to identify the dead zone interval, the high friction interval and the stroke boundary in time, and the diagnostic results are mostly in the form of alarms or maintenance suggestions, which are not closely coupled with the process control system, and the control command is difficult to avoid the known high-risk stroke interval in time.

[0004] Under the above engineering conditions and application constraints, how to obtain segmented parameter information sufficient to reflect the mechanical and response characteristics of different stroke intervals by using only a small amount of external measurable signals such as driving amount and valve stem displacement, and to construct safety operation constraints that can be directly used by the process control system on the inventory valves lacking original design data, becomes a core technical problem to be solved under the premise of not stopping production or stopping production as little as possible and keeping the valve structure unchanged.

[0005] To solve the above problems, a technical solution is provided. SUMMARY

[0006] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present application provide a valve detection control method based on front and rear ends, which performs a small stroke exploration at the front end terminal, synchronously collects the valve rod angular displacement and the actuator driving amount under a unified time reference, obtains the stroke segmentation and the segment-level characteristic parameter set through incremental segmentation fitting, models the segment-level characteristic parameter set and the process constraint data in the stroke segmentation dimension by the rear end platform to generate a health constraint parameter set and issues it to the front end terminal, and the front end terminal projects the target opening degree instruction according to the health constraint parameter set, limits the action speed, and continuously monitors the angle deviation, retreats control based on the historical safe position, and reports abnormal events, so as to realize the integrated management of the online diagnosis result and the control behavior of the inventory valve, thereby solving the problems proposed in the background art.

[0007] To achieve the above object, the present application provides the following technical scheme: S1: after the front end terminal is powered on, self-checking and zero calibration are performed, the valve rod is driven to reciprocate and micro-move in a preset exploration sequence within a local small stroke of the valve, the valve rod angular displacement and the actuator driving amount are synchronously collected under a unified time reference and stored in time sequence to form small stroke exploration data; S2: the front end terminal pre-processes the small stroke exploration data according to a unified time step, calculates the driving amount and displacement increment of adjacent sampling points, and performs change point detection and segmentation fitting to obtain a plurality of stroke segments on the valve rod stroke, calculates the segment-level characteristic parameters reflecting friction and hysteresis for each stroke segment, generates a segment-level characteristic parameter set, and uploads it to the rear end platform; S3: the rear end platform receives the segment-level characteristic parameter set, and correspondingly organizes the process constraint data of the valve loop in the stroke segmentation dimension, calculates the friction characteristics and flow capacity of each stroke segment based on a simplified working condition model, obtains the safe opening degree upper limit and the maximum allowable action speed of each stroke segment, forms a health constraint parameter set, and issues it to the front end terminal; S4: when receiving the target opening degree instruction, the front end terminal converts the target opening degree into a target angle and locates the target angle on the stroke segment, corrects the target angle and action speed according to the health constraint parameter set to generate an angle trajectory, monitors the angle deviation during execution, and when the angle deviation continuously exceeds the set standard, retreats control according to the last safe position and reports abnormal events to the rear end platform.

[0008] Further, step S1 comprises: the front-end terminal drives the valve stem to reciprocate and micro-move in a local range of the full stroke of the valve with the current working position as the center in a way of gradually increasing the stroke amplitude when generating the preset trial sequence, and after collecting the angular displacement of the valve stem and the driving amount of the actuator, the small-stroke trial data with the driving amount and displacement change below the threshold value are removed, and the small-stroke trial data used for stroke segmentation processing are reserved.

[0009] Further, step S1 further comprises: after completing the zero position calibration, the front-end terminal determines the reference point of the angle sensor by using the mechanical limit or small-amplitude reciprocating rotation of the clamping mechanism, and assigns a unique trial number to each trial action under a unified time reference, records the trial action direction, trial start time and trial end time, and stores the small-stroke trial data in the front-end local buffer.

[0010] Further, step S2 comprises: when the front-end terminal pre-processes the small-stroke trial data, the angular displacement of the valve stem and the driving amount of the actuator are resampled according to a unified time step, high-frequency noise is weakened through digital filtering, and then the resampled results are subjected to amplitude normalization processing, and then the driving amount increment and displacement increment of adjacent sampling points are calculated to construct an increment sequence for change point detection.

[0011] Further, step S2 further comprises: after completing the change point detection and segmentation fitting, the front-end terminal records the plurality of stroke segments obtained by dividing the valve stem stroke in the stroke segment division table according to the stroke segment number, and writes the average slope, static friction peak value, hysteresis width and stroke segment type label corresponding to each stroke segment into the segment-level characteristic parameter set, and stores the stroke segment number as the index and uploads it to the back-end platform.

[0012] Further, step S3 comprises: the process constraint data at least includes the upstream pressure allowable range, the downstream pressure allowable range, the upper limit of the allowable pressure difference, the target flow allowable range and the medium density of the valve belonging to the loop, and the back-end platform writes the segment-level characteristic parameter set and the process constraint data into the same joint data structure in the stroke segment dimension, which is used to construct a simplified working condition model.

[0013] Further, step S3 further comprises: based on the joint data structure, the back-end platform calculates a health index representing the friction state and driving force demand for each stroke segment, and according to a pre-established threshold interval, divides the risk level of the stroke segment, and the health index and the risk level together with the stroke segment safety opening degree upper limit and the stroke segment maximum allowable action speed form the stroke segment record in the health constraint parameter set.

[0014] Further, step S4 comprises: the front-end terminal first projects the target angle on all travel segments involved in the motion path to obtain a safe target angle, and then selects a path maximum motion speed from the maximum allowed motion speeds of each travel segment, and constructs a planning angle trajectory composed of an acceleration segment, a constant speed segment and a deceleration segment according to the safe target angle, the path maximum motion speed and the sampling period when generating the angle trajectory according to the health constraint parameter set.

[0015] Further, step S4 further comprises: the front-end terminal performs multiple reference motion tests on the health valve in advance when monitoring the angle deviation, and statistics the distribution of the absolute value of the angle deviation at each sampling time, determines the angle deviation threshold according to the distribution, and statistics the distribution of the number of continuous overruns in the health motion, determines the number of continuous overruns standard according to the distribution, and uses the angle deviation threshold and the number of continuous overruns standard as the abnormal triggering condition in online operation.

[0016] Further, step S4 further comprises: the front-end terminal records the actual angle position at which the last motion ends and the angle deviation does not exceed the angle deviation threshold as the last safe position when the angle deviation continuously exceeds the threshold and triggers the rollback control, uses the last safe position as the rollback end point in the rollback control process, and generates an abnormal event record after the rollback ends, the abnormal event record at least includes the original target opening instruction, the safe target angle, the set of motion path travel segments, the maximum angle deviation, the rollback starting angle position and the rollback termination angle position, and is uploaded to the back-end platform.

[0017] The valve detection and control method based on the front-end and back-end has the following technical effects and advantages: The present application unifies the friction characteristics, hysteresis characteristics of the valve stem in different angle intervals and process parameters such as loop pressure and flow into the same travel segment coordinate system through the cooperation of front-end small travel exploration, travel segment modeling and back-end working condition constraint calculation, forms a health constraint parameter set with clear angle boundary, and is delivered to the front-end terminal. When the front-end executes the target opening instruction, it no longer only acts according to a single set value, but first projects the target angle on the travel segments, and then generates an angle trajectory combined with the maximum allowed motion speed of each travel segment on the path, directly converts the diagnosis result into an executable control constraint, and realizes a coherent chain from data acquisition, state identification to control instruction landing.

[0018] Meanwhile, the application introduces angle deviation continuous monitoring and automatic back-off mechanism based on historical safe position in front-end terminal. When abnormal trend such as blockage and step loss occurs in a stroke segment, the original planned trajectory can be terminated and back-off to the verified safe angle position in local control layer, and the abnormal event record containing target instruction, stroke segment information and deviation statistics is uploaded to the back-end. Through the combination of segment health constraint and abnormal closed-loop reporting, the application can fine manage the aging valve and the inventory valve under complex working conditions in a unified method without interrupting production and relying on frequent disassembly and inspection, so that the valve can keep in the safe working interval matching its own state during long-term operation. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 The flowchart of the valve detection and control method based on the front-end and back-end of the application is shown. DETAILED DESCRIPTION

[0020] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, not all. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the application.

[0021] Embodiment 1: Figure 1 The valve detection and control method based on the front-end and back-end of the application is given, including: S1: After the front-end terminal is powered on, self-checking and zero calibration are performed, the valve rod is driven to reciprocate and move in a small stroke according to a preset trial sequence, the valve rod angular displacement and actuator driving amount are synchronously collected under a unified time reference and stored in time sequence to form small stroke trial data; S2: The front-end terminal pre-processes the small stroke trial data according to a unified time step, calculates the driving amount and displacement increment of adjacent sampling points, performs change point detection and segment fitting, obtains a plurality of stroke segments on the valve rod stroke, calculates segment-level characteristic parameters reflecting friction and hysteresis for each stroke segment, generates a segment-level characteristic parameter set and uploads it to the back-end platform; S3: The back-end platform receives the segment-level characteristic parameter set, and correspondingly organizes the process constraint data of the valve belonging to the loop in the stroke segment dimension, calculates the friction characteristics and flow capacity of each stroke segment based on a simplified working condition model, obtains the safe opening upper limit and maximum allowable action speed of each stroke segment, forms a health constraint parameter set and issues it to the front-end terminal; S4: When the target opening degree instruction is received, the front-end terminal converts the target opening degree into a target angle and positions the target angle on the stroke segment, corrects the target angle and the action speed according to the health constraint parameter set to generate an angle trajectory, monitors the angle deviation in the execution process, and performs a rollback control according to the last safe position and reports an abnormal event to the back-end platform when the angle deviation continuously exceeds a set standard.

[0022] In the use scenario of the inventory valve under long-term pressure operation, the primary work is to obtain original data that can reflect the motion state of the valve stem, and at the same time cannot cause significant working condition disturbance. Step S1 focuses on the front-end terminal to complete static detection, zero-position angle reference establishment, small-stroke trial target generation, and synchronous sampling and abnormality screening outside the original valve, aiming to output a small-stroke trial data set established on the basis of unified time reference and unified angle reference. The data set will be the only input source for driving incremental and angle incremental relationship identification in the subsequent steps, so in this step, each data object needs to be given a unique name and fixed usage.

[0023] S101. Self-checking and static data acquisition.

[0024] After the front-end terminal is powered on, the valve stem is kept in a mechanical position unchanged, and the output of the angle sensor is read at a fixed sampling period within a preset static detection time interval to form a static angle sampling sequence arranged in the order of sampling time. The difference between the maximum angle sampling value and the minimum angle sampling value in the static angle sampling sequence defines an angle static fluctuation value. The angle static fluctuation value is used to measure the stability of the angle sampling of the valve stem under static conditions.

[0025] Within the same static detection time interval, the output of the driving acquisition unit is read at the same sampling period to form a static driving sampling sequence. The difference between the maximum driving sampling value and the minimum driving sampling value in the static driving sampling sequence defines a driving static fluctuation value. The angle static fluctuation value and the driving static fluctuation value are compared with the preset angle stability threshold and driving stability threshold, respectively.

[0026] When the angle static fluctuation value does not exceed the angle stability threshold and the driving static fluctuation value does not exceed the driving stability threshold, it is determined that the static detection is passed, otherwise it is determined that the sensor or the actuator is in an abnormal state and the subsequent steps are stopped.

[0027] S102. Zero reference angle calibration.

[0028] After the static detection is passed, a reciprocating rotation instruction of a pre-defined angle range is sent to the actuator, so that the valve stem performs a forward rotation and a reverse rotation around the current working condition position. During the entire reciprocating rotation process, the angle sensor output is continuously read at the aforementioned sampling period to form a dynamic angle sampling sequence.

[0029] The minimum and maximum angle sampling values appearing in the dynamic angle sampling sequence represent the two angle limits reached by the current reciprocating action. The arithmetic mean of the two angle limits is defined as the zero reference angle value. The zero reference angle value serves as the only reference for the valve stem angle in the entire invention, and all subsequent angle-related data is expressed in the form of deviation from the zero reference angle value. To ensure consistency between different valves, the angle range of reciprocating rotation is set during the installation and debugging stage according to the maximum exploration amplitude allowed by the valve under low load conditions. Typically, by gradually increasing the reciprocating command amplitude and observing pressure fluctuations and flow fluctuations, the maximum amplitude that does not cause significant operating condition changes is selected as the configuration value before triggering an alarm.

[0030] S103. Small stroke exploration target generation.

[0031] After the zero reference angle value is established, the current angle sensor output is read, and the angle sampling value is subtracted from the zero reference angle value to obtain the current operating condition relative angle value. A dimensionless small stroke increment list is stored internally, and each increment unit in the list represents the expected angle amplitude offset relative to the current operating condition relative angle value, and the increment list is monotonically increasing in value. For each increment unit, two types of target relative angle values are constructed, one is the current operating condition relative angle value plus the increment unit, corresponding to the positive exploration target angle value, and the other is the current operating condition relative angle value minus the increment unit, corresponding to the negative exploration target angle value. Each target relative angle value is added to the zero reference angle value to obtain the positive exploration target absolute angle value and the negative exploration target absolute angle value. Each pair of positive and negative exploration target absolute angle values is assigned a unique exploration number, and the exploration numbers are arranged in increasing order according to the increment list to form a complete exploration target angle queue. To ensure that the exploration action does not exceed the small stroke range allowed by the valve, the increment list is determined during the configuration stage through multiple tests, specifically by gradually increasing the relative angle offset amplitude, and selecting a maximum offset amplitude value under the premise that no alarm occurs in monitoring the supply pressure and flow changes, and then dividing the range from zero to the maximum offset amplitude value into several levels to form the small stroke increment list.

[0032] S104. Small stroke exploration execution and synchronous sampling.

[0033] According to the order of the trial target angle queue, the trial action is executed from the entry with the smallest trial number. For each trial number, a motion trajectory composed of angular velocity is planned, which is divided into three stages in time, i.e. an acceleration stage with gradually increasing angular velocity, a constant velocity stage with fixed angular velocity, and a deceleration stage with gradually decreasing angular velocity. The time length of the three stages and the angular velocity variation rate are configured by the control engineering technician according to the inertia of the actuator and the mechanical characteristics of the valve during installation and debugging. The actuator drives the valve stem along the motion trajectory to run from the initial angle position to the corresponding trial target absolute angle value and return to the initial position.

[0034] During the entire action process, the angle sensor output and the drive acquisition unit output are synchronously read at a fixed sampling period. The angle sampling value, the drive sampling value obtained each time the sampling is performed, and the absolute time at which the sampling occurs are combined to form a data unit, which is stored in the data sequence corresponding to the current trial number in the order of increasing sampling sequence number from one.

[0035] After the action execution of all trial numbers, a plurality of data sequences are formed in the internal storage area of the front-end terminal, each data sequence corresponding to a trial number and internally containing angle sampling values, drive sampling values, and sampling times arranged in time sequence.

[0036] S105. Increment calculation and abnormality screening.

[0037] In order to provide a discrete correspondence between drive change and angle change for subsequent steps, in the data sequence of each trial number, two adjacent data units are sequentially taken according to the sampling sequence number, the angle sampling value of the latter data unit is subtracted from the angle sampling value of the former data unit to obtain a series of angle increment values, which constitute an angle increment sequence; at the same time, the drive sampling value of the latter data unit is subtracted from the drive sampling value of the former data unit to obtain a series of drive increment values, which constitute a drive increment sequence.

[0038] The sum of the absolute values of all angle increment values of each angle increment sequence is obtained, which is the total angle change corresponding to the trial number. The sum of the absolute values of all drive increment values of each drive increment sequence is obtained, which is the total drive change corresponding to the trial number. The total angle change is compared with the preset minimum angle change threshold value. When the total angle change is less than the minimum angle change threshold value, it is considered that no effective angle displacement is generated in this trial, and this trial is identified as an idle trial. The total drive change is compared with the preset maximum drive change threshold value. When the total drive change is greater than the maximum drive change threshold value, it is considered that the actuator is close to a saturated operating state in this trial, and the trial data is not suitable for characteristic modeling.

[0039] The above screening logic is applied to all trial numbers one by one, and the data sequence determined as an idle trial or a nearly saturated trial is marked as an abnormal trial in whole and removed from the modeling data set, leaving only the data sequence corresponding to the trial number that passes the screening.

[0040] After the implementation of step S1, a qualified trial data set has been formed in the front-end terminal, which is indexed by trial number as outer index and by sample sequence number as inner index, uses zero-bit reference angle value as angle reference uniformly, contains static detection results, small-stroke trial target angle information, angle sample values, drive sample values, sampling times, and angle total change and drive total change calculated from these sample values, and makes clear marking for abnormal trials. This data set is directly used in subsequent step S2 to establish a segmented correspondence between drive increment and angle increment, thereby completing the identification of valve stroke segmentation and high-risk interval under the condition of unknown original design parameters, and providing a solid data foundation for subsequent health constraint generation.

[0041] In one embodiment, the regulating valve of a certain chemical device exhibits slow opening command to position after long-term operation. The maintenance personnel fix the front-end terminal on the clamping mechanism near the valve stem and connect the angle sensor and the drive amount collection line of the actuator to the front-end terminal without shutdown. After the front-end terminal is powered on, self-checking is completed, and then the drive actuator is driven to make a small amplitude reciprocating rotation. The clamping mechanism forms a clear reference point at the mechanical limit position, and the front-end terminal completes zero calibration and aligns the angle sampling and drive amount sampling to the same time reference accordingly. After the maintenance personnel confirm that the valve is near the opening degree allowed by the process on site, the front-end terminal drives the valve stem to reciprocate around the current working condition position according to the small-stroke trial sequence, and the front-end terminal synchronously collects the valve stem angular displacement and actuator drive amount and writes them into the local buffer area in time sequence, while writing the trial number, action direction, and start and end time for each trial action. After the trial is completed, the front-end terminal locally marks the segment with nearly static angular displacement change as invalid and does not enter subsequent analysis. On site, the front-end terminal can be seen to generate a set of trial data files arranged by trial number and enter the state of being analyzed.

[0042] The qualified test data set is obtained in step S1, and the angle sampling value, driving sampling value and sampling time of each small stroke test are recorded in the set according to the test number and sampling sequence number, and the idle test and saturated test are removed. For the inventory valve, in order to extract the stroke segment characteristics for the back-end platform from these discrete time series, the original data of different tests and different time steps need to be unified to a common reference frame, and the stroke segment with clear physical meaning in the angle coordinate is identified. Step S2 is carried out around this goal, and the test data is converted into segment-level characteristic parameters indexed by angle position through time resampling, increment construction, gradient analysis, single test stroke segment statistics and cross-test aggregation, to provide a basis for step S3 of working condition modeling and health constraint generation.

[0043] S201. Test data inheritance and standard time resampling.

[0044] When starting processing, the data is read in the order of test number from the qualified test data set formed in step S1. For any test number, the data contains an angle sampling value sequence, a driving sampling value sequence and a sampling time sequence arranged according to the sampling sequence number. Due to the influence of actuator response and communication time delay in the actual sampling process, the adjacent sampling time intervals in the same test may be different, and the sampling time intervals between different tests are also inconsistent, so it is necessary to establish a standard time sequence within each test number.

[0045] The specific method is to first extract the first sampling time and the last sampling time of the current test, divide this time interval into several continuous time points according to the pre-set standard time step, and form a standard time sequence. The standard time step can be determined according to the maximum allowed angular velocity of the actuator and the desired time resolution during installation and debugging, for example, a step length that contains at least several tens of standard time points in one test is selected. For each time point in the standard time sequence, find the two nearest sampling time points in the original sampling time sequence, assume that the angle sampling value and the driving sampling value change at a constant speed between the two sampling time points, and calculate the resampled angle value and the resampled driving value corresponding to the standard time point according to the linear interpolation method. The meaning of interpolation calculation is that the angle value of the standard time point between the two actual sampling time points is equal to the angle value at the previous time plus the product of the angle change rate and the time offset, and the driving value is the same. Through this resampling process, each test number obtains a set of resampled angle value sequence and resampled driving value sequence defined on the standard time sequence, thereby establishing a unified time reference between different tests.

[0046] S202. Increment construction and gradient ratio calculation.

[0047] After standard-time resampling, angle increments and driving increments are constructed for the resampled angle value sequence and resampled driving value sequence within each trial number. For any two adjacent standard time points within a trial number, the angle increment is defined as the resampled angle value of the later standard time point minus the resampled angle value of the earlier standard time point, and the driving increment is defined as the resampled driving value of the later standard time point minus the resampled driving value of the earlier standard time point. The resulting sequence of angle increments and driving increments reflects the angle change and driving change within each time step during a trial.

[0048] To describe the response strength of driving changes to angle changes from a local perspective, the gradient ratio is calculated for each time step, provided the absolute value of the angle increment is higher than the angle increment filtering threshold. The corresponding driving increment value is then divided by the angle increment value to obtain the driving change per unit angle change. For time steps where the absolute value of the angle increment is not higher than the angle increment filtering threshold, the gradient ratio is not calculated; instead, these time steps are marked as angle stagnation points in the data record, indicating that the angle has not changed significantly within this time period.

[0049] For example, the gradient ratio can be calculated as follows: After the unified time-based re-sample collection is completed, the control unit will handle any probe number. Two adjacent time indices inside and The angle increment is defined as the subsequent resampled angle value minus the previous resampled angle value, and the drive increment is defined as the subsequent resampled drive value minus the previous resampled drive value. The expression is as follows: ; ; In the formula, the angle increment Indicates the number of the trial. In time index The angle change between the current time index and the next time index drives the increment. This represents the driving change within the same time interval. To reflect the response strength of the driving change to the angle change from a local perspective, the control unit divides the driving increment by the angle increment to obtain the gradient ratio, provided that the absolute value of the angle increment is greater than the angle increment filtering threshold. The expression is: ; gradient ratio in the formula Indicates the number of trials Time Index Nearby, the unit angle change corresponds to the driving change amount. The angle increment filtering threshold is used to distinguish the sampling time when the angle is basically static and the sampling time when there is actual motion. In the field debugging stage, a stable test data can be selected, the absolute values of all angle increments are counted, a representative value near the lower limit of the statistical distribution is selected and multiplied by a proportion coefficient less than one to obtain.

[0050] In one embodiment, the maintenance personnel installs the front terminal on site before the valve is installed, and completes the short-range test of step S1. The front terminal obtains the sampling time sequence, the angle sampling value sequence and the driving sampling value sequence corresponding to the same test number during the test. The front terminal first generates a standard time sequence according to the standard time step, and then generates a resampled angle value sequence and a resampled driving value sequence at each standard time point by linear interpolation. Then, the front terminal calculates the angle increment value and the driving increment value pair by pair according to the adjacent standard time points, and compares the angle increment value with the angle increment filtering threshold. The standard time step that meets the filtering condition is written into the gradient ratio value, and the standard time step that does not meet the filtering condition is written into the angle stagnation point mark and synchronously records the stagnation driving increment table. After the conversion is completed, the front terminal obtains the gradient ratio sequence and the angle stagnation point index set corresponding to the original test number one by one, which are directly input for the gradient change amount calculation of step S203.

[0051] S203. Gradient change amount calculation and single test candidate boundary identification.

[0052] In order to identify the mutation position of the driving response characteristic to the angle in each test record, the gradient change amount is further calculated on the obtained gradient ratio sequence. The gradient change amount is defined as the absolute value of the difference between the gradient ratios of two adjacent time steps, and its physical meaning is the change amplitude of the driving response intensity to the angle per unit time step.

[0053] For each test number, the absolute value of the difference between adjacent gradient ratios is calculated one by one from the first defined gradient ratio. These gradient change amounts are compared with the gradient change amount judgment threshold. The time index whose gradient change amount is higher than the judgment threshold is collected into the candidate boundary index set.

[0054] Since the candidate boundary index set often contains a string of time indexes close to each other, in order to reduce the boundary redundancy in the angle coordinate, the candidate boundary indexes are clustered and compared. When the difference between two candidate indexes is not higher than the preset index interval threshold, the two indexes are classified into the same change band. The time index with the highest gradient change amount value in the change band is selected as the representative boundary index. In each test record, the representative boundary index, the starting time point index and the ending time point index together constitute the time boundary set of a single test, and the corresponding resampled angle value forms the single test angle boundary set.

[0055] S204. Single-probe stroke segment division and segment-level property calculation.

[0056] After obtaining the single-probe angle boundary set, all time indices between two adjacent time boundaries are divided into a single-probe stroke segment in time sequence. Each single-probe stroke segment has a corresponding start angle position and end angle position, as well as an angle increment value sequence, a driving increment value sequence and a gradient ratio sequence covering the segment. The average gradient value is first calculated within each single-probe stroke segment, which is obtained by summing all defined gradient ratios in the segment and dividing the sum by the number of gradient ratios in the segment, representing the average stiffness level of the whole segment. Then the absolute maximum value of driving increment value is found within the segment, which is taken as the driving peak increment to reflect the peak characteristics of driving change in the segment. At the same time, the angle net change is calculated, which is obtained by accumulating all angle increment values within the segment, representing the total angle change from the start point to the end point of the segment.

[0057] The ratio of the number of time indices marked as angle stagnation points to the total number of time indices in the segment can also be calculated to obtain the angle stagnation proportion, which is used to reflect the degree of long-term angle inactivity in the segment. Based on the combination of average gradient value, driving peak increment, angle net change and angle stagnation proportion, the type of each single-probe stroke segment is determined. When the angle net change is not higher than the preset dead zone angle threshold, and the angle stagnation proportion is high, the segment is marked as a dead zone segment; when the angle net change is higher than the dead zone angle threshold, the average gradient value is higher than the upper limit of the normal gradient reference range, and the driving peak increment is higher than the static friction judgment threshold, the segment is marked as a high friction segment; other segments are marked as normal segments.

[0058] Each single-probe stroke segment finally forms a single-probe segment-level property record containing the start angle position, the end angle position, the average gradient value, the driving peak increment, the angle net change, the angle stagnation proportion and the segment type label, and is stored in the intermediate property data set with the probe number and the segment number as the index.

[0059] S205. Cross-probe angle boundary aggregation and global stroke segment property generation.

[0060] Since the start angle position and the probe amplitude of each probe may be different, the single-probe stroke segments formed in different probes overlap and interleave in the angle coordinate.

[0061] In order to establish uniform stroke segment division in the full valve stroke range, the segment start angle position and the segment end angle position are extracted from all single trial segment level characteristic records, and the two types of angle values are integrated into a set of original boundary angles. After the set of original boundary angles is sorted in ascending order of angle value, the difference between the two adjacent boundary angles is compared with the global angle merging threshold. For a group of continuous boundary angles with a difference not higher than the global angle merging threshold, they are classified into the same angle cluster, and the arithmetic mean of all angle values in the angle cluster is calculated as a global stroke boundary angle. By performing this process on the entire sequence of original boundary angles, a number of global stroke boundary angles arranged in ascending order of angle are formed, and the angle interval between any two adjacent global stroke boundary angles defines a global stroke segment.

[0062] After the global stroke segments are divided, for each global stroke segment, all single trial segment level characteristic records are traversed to collect those records whose start angle position and end angle position completely fall within the current global stroke segment angle range to form a segment statistics set. For each record in the statistics set, the average gradient value, the driving peak increment, and the angle net change are read, and the arithmetic mean of each type of characteristic is calculated to obtain the global average gradient, the global average driving peak, and the global average angle net change of this global stroke segment. The segment type labels in the records can also be counted to count the proportion of the number of dead zone segments, high friction segments, and normal segments, which are used to verify the consistency of global indicators and local types.

[0063] Then, according to the relationship between the global average gradient, the global average driving peak, the global average angle net change, and the aforementioned thresholds, a final stroke segment type label is assigned to each global stroke segment. When the global average angle net change is within the dead zone angle threshold range and the proportion of the number of dead zone segments in the statistics set is the highest, the global stroke segment is identified as a dead zone segment; when the global average gradient and the global average driving peak are both higher than the upper limit of the normal reference range and the proportion of the number of high friction segments is the highest, the global stroke segment is identified as a high friction segment; otherwise, it is identified as a normal segment. The final segment level characteristic parameter set is indexed by the global stroke segment number, records the start angle position, the end angle position, the global average gradient, the global average driving peak, the global average angle net change, and the stroke segment type label of each segment, and is stored in the front-end non-volatile storage and sent to the back-end platform through the communication link, providing direct input for the working condition model construction and health constraint parameter calculation in step three.

[0064] After the processing of step S2, the small stroke test timing data from step S1 is completely transformed into stroke segment characteristic description with physical meaning in angle coordinate. Each global stroke segment corresponds to a specific angle interval of the valve stem, which contains statistical average gradient and driving peak information from multiple tests, as well as segment type label determined by threshold judgment. In this way, the back-end platform can directly superimpose pressure and flow constraints on specific stroke segments in step S3 to generate safe opening interval and action speed limit that meet the existing working conditions, so that health constraints no longer remain at the overall valve level, but are implemented to the fine angle interval.

[0065] In one embodiment, after the front-end terminal completes step S1, the maintenance personnel do not disassemble the front-end terminal on site, but directly start the stroke segment identification through the maintenance interface of the front-end terminal. The front-end terminal first re-aligns the angle displacement sequence corresponding to each test number with the driving amount sequence according to a uniform time step to form a time sequence that can be directly compared, and then performs denoising processing on the sequence to make the mutation points caused by sensor jitter not participate in the segment judgment. The front-end terminal then calculates the driving amount increment and displacement increment of adjacent time points point by point, and forms an increment sequence arranged by time. The front-end terminal performs change point detection on the increment sequence, and the change point position corresponds to the time when the valve stem response law changes. Then, the front-end terminal performs segment fitting on each segment of data with the change point as the boundary to obtain multiple stroke segments. The front-end terminal writes the starting angle position, ending angle position, average slope, static friction peak value, hysteresis width and stroke segment type label of each stroke segment into the segment-level characteristic parameter set, and forms a stroke segment division table indexed by stroke segment number. The maintenance personnel can see on the interface that the stroke segment number and the segment boundary angle list have been generated, and then the front-end terminal packages the segment-level characteristic parameter set and enters the upload queue.

[0066] In the foregoing steps, the front-end has formed a segment-level characteristic parameter set through small stroke test and segment identification, and uploaded it to the back-end platform. The segment-level characteristic parameter set is indexed by stroke segment number, and records the starting angle position, ending angle position, global average gradient, global average driving peak value, global average angle net change, and stroke segment type label segment by segment. For the inventory valves in complex process circuits, it is not enough to directly determine which stroke segments can be safely used under the current upstream pressure, downstream pressure, allowable pressure difference and target flow range by relying only on these mechanical characteristic parameters. The back-end platform needs to introduce process constraint data, unify mechanical characteristics and working condition constraints in one data structure by one-to-one correspondence with the segment-level characteristic parameters in the stroke segment number dimension, and then calculate the theoretical flow capacity, health index, safe opening upper limit and maximum allowable action speed of the stroke segment on this basis to form a health constraint parameter set that can be executed by the front-end.

[0067] S301. Correspondence establishment and arrangement between segment-level characteristic parameters and process constraint data.

[0068] The backend data processing unit first reads the aforementioned segment-level characteristic parameter set. For each stroke segment number, the data processing unit takes out the starting angle position, the ending angle position, the global average gradient, the global average driving peak value, the global average angle net change amount, and the stroke segment type label of the segment one by one. Subsequently, the data processing unit reads the process constraint data corresponding to the loop to which the valve belongs from the process monitoring database according to the valve unique identifier, and the process constraint data includes the upstream pressure allowable minimum value, the upstream pressure allowable maximum value, the downstream pressure allowable minimum value, the downstream pressure allowable maximum value, the allowable pressure difference upper limit, the target flow allowable minimum value, the target flow allowable maximum value, and the medium density. The data processing unit constructs a joint data table with the stroke segment number as the primary key, writes the segment-level characteristic field of each stroke segment and the aforementioned process constraint field into the same row record, so that the stroke segment number and the corresponding mechanical parameters and working condition parameters are completely corresponding in structure. If it is found that the pressure unit, flow unit or angle unit in the process constraint data is inconsistent with the unit adopted by the segment-level characteristic parameter, the data processing unit will convert all pressure values into one pressure unit, all flow values into one flow unit, and all angle values into one angle unit according to the pre-set unit conversion table, and then write them into the joint data table.

[0069] Through this way of establishing a joint data table according to the stroke segment number, the segment-level characteristic parameter set and the process constraint data complete the segment-by-segment correspondence and arrangement at the data level.

[0070] S302. Calculation logic of stroke segment effective pressure difference and theoretical flow capacity.

[0071] After the joint data table is established, the modeling unit calculates the effective pressure difference and the theoretical flow capacity for each stroke segment using the content in the table. For any stroke segment number, the modeling unit first reads the maximum upstream pressure allowance and the minimum downstream pressure allowance from the joint data table, calculates the difference between the two as the theoretical pressure difference upper limit. Then compare the theoretical pressure difference upper limit with the allowable pressure difference upper limit, take the smaller one of the two as the effective pressure difference available for the stroke segment under the current working condition. Subsequently, the global average gradient needs to be converted into an equivalent flow coefficient. To this end, during the offline test phase, select a number of reference valves similar to the target valve specification, perform standardized flow tests on these reference valves according to fixed stroke segment division, collect stable flow and pressure difference data at different segment angle positions, calculate the flow coefficient of each reference stroke segment through traditional flow coefficient definition, and record the corresponding global average gradient, finally form a set of discrete corresponding points between global average gradient and flow coefficient. The modeling unit performs interpolation fitting on this set of corresponding points to obtain the functional relationship from global average gradient to equivalent flow coefficient, and stores the function parameters in the configuration file.

[0072] During online operation, for each stroke segment in the joint data table, the modeling unit reads the global average gradient, calls the aforementioned functional relationship, and calculates the equivalent flow coefficient through interpolation operation. Then according to the conventional calculation method of orifice type flow, multiply the equivalent flow coefficient of the stroke segment by the square root of the ratio of effective pressure difference to medium density to obtain the theoretical maximum flow capacity corresponding to the stroke segment per unit time. By appending a column of theoretical maximum flow capacity field in the joint data table, execute the above steps for all stroke segments to establish the corresponding relationship between stroke segments and theoretical flow capacity.

[0073] For example, the conversion of global average gradient to equivalent flow coefficient can be processed as follows: Let the discrete values of global average gradient be sorted in ascending order as ; The corresponding flow coefficient calibration values are sorted as ; When the global average gradient value falls into the interval , the equivalent flow coefficient ; ; When is less than , the equivalent flow coefficient is ; when is greater than , the equivalent flow coefficient is . ​

[0074] The breakpoint values of the global average gradient are sorted from small to large, and the total number of breakpoints is The breakpoint sequence of When the global average gradient is located between the The linear interpolation of the The value of ranges from 2 to -1 。

[0075] In an embodiment, the tester sets up a closed-loop pipeline at the test station and installs a reference valve, clamps the front-end terminal at the valve stem position of the reference valve, and completes the small-stroke exploration and stroke segmentation identification according to steps S1 and S2 to obtain the global average gradient corresponding to each stroke segmentation number. The tester then drives the valve stem to the midpoint angle position of a stroke segmentation and keeps it stationary, adjusts the return branch to make the flowmeter reading enter a stable state, and records the flowmeter reading and upstream and downstream pressure measurement point readings in the stable sampling window, calculates the differential pressure, and obtains the flow coefficient calibration value of the stroke segmentation according to the definition of the flow coefficient. The tester repeats the above-mentioned keeping and recording process for each stroke segmentation to form a discrete corresponding point set of the global average gradient and the flow coefficient calibration value. The back-end platform sorts the discrete corresponding points according to the global average gradient and generates a breakpoint array, writes it into a configuration file, and calculates the equivalent flow coefficient by linear interpolation during online operation.

[0076] S303. Stroke segmentation health index construction and risk level division logic.

[0077] In order to reflect the comprehensive situation of friction state and driving load at the stroke segmentation level, the modeling unit constructs a stroke segmentation health index based on the global average gradient and the global average driving peak value. Before formal operation, the maintenance personnel collect two types of segment-level characteristic samples on a number of valves, one type is a healthy valve sample with good structure and lubrication state, and the other type is a fault valve sample with obvious jamming, wear or leakage. For healthy valve samples, the global average gradient of all segments is counted, and the central distribution position is taken as the gradient reference value; the global average driving peak value of all segments is counted, and the central distribution position is taken as the driving reference value.

[0078] The centralized distribution position defining method is as follows: the global average gradient of all travel segments in the health valve sample is used to form a sample set, the sample set is sorted in ascending order of numerical value, the samples at both ends of the sorted sequence are removed by a preset proportion, the middle sample subset is retained, and the arithmetic mean of the middle sample subset is taken to obtain the gradient reference value. The global average driving peak value of all travel segments in the health valve sample is used to form another sample set, and the same sorting, removal and arithmetic mean process is used to obtain the driving reference value. The preset proportion is given by the maintenance regulation, and the setting principle is to remove a small number of abnormal segments that deviate obviously, while retaining enough normal segments for calculating the reference value.

[0079] In one embodiment, the maintenance personnel select several valves with good structure and lubrication state in the same factory area as health valve samples, install the front-end terminal one by one, and perform steps S1 and S2, and upload the segment-level characteristic parameter set of each valve to the back-end platform. The back-end platform extracts the global average gradient from the segment-level characteristic parameter set of each health valve sample and combines it into a gradient sample set, sorts the gradient sample set and removes the samples at both ends by a preset proportion, obtains the middle sample subset and calculates its arithmetic mean, and writes it into the gradient reference value field. The back-end platform processes the global average driving peak value in the same way to obtain the driving reference value field. The back-end platform reads the global average gradient and the global average driving peak value for any travel segment during online operation, and performs ratio calculation with the gradient reference value and the driving reference value respectively, as input for subsequent health index construction.

[0080] During online operation, the modeling unit reads the global average gradient and the global average driving peak value for any travel segment number from the joint data table, divides them by the gradient reference value and the driving reference value respectively to obtain the average gradient ratio and the driving peak value ratio. Then multiply the average gradient ratio and the driving peak value ratio to obtain a comprehensive resistance factor, and divide one by one plus the comprehensive resistance factor to obtain the health index value. The health index naturally falls between zero and one, the closer the value is to one, the closer the friction state and driving load of the segment are to the reference level of the health valve, and the closer the value is to zero, the closer the segment is to a serious resistance state.

[0081] To convert the health index into discrete risk levels, the modeling unit determines two health index thresholds using the health index distribution of the aforementioned health valve samples and failure valve samples. The specific process is as follows: sort the segment health indexes of all health samples, select the values at the lower preset proportion and the upper preset proportion of the sorted sequence as the interval boundaries of the health samples; similarly, sort the segment health indexes of all failure samples, select the values at the lower preset proportion and the upper preset proportion of the sorted sequence as the interval boundaries of the failure samples. Then, by comparing the interval boundaries of the two types of samples, the low-risk, medium-risk and high-risk zones on the health index axis are determined.

[0082] When running online, for each stroke segment number, the modeling unit divides the risk level according to the interval where the health index falls and in combination with the stroke segment type label. When the health index falls in the low-risk interval and the stroke segment type label is normal segment, the risk level is recorded as low risk; when the health index falls in the medium-risk interval or the stroke segment type label is normal segment but the health index is close to the high-risk boundary, the risk level is recorded as medium risk; when the health index falls in the high-risk interval or the stroke segment type label is dead zone segment or high friction segment, the risk level is recorded as high risk.

[0083] The risk level field is written back to the joint data table, so that each stroke segment has both a health index and a risk level.

[0084] S304. Safety opening upper limit and maximum allowable action speed solving and health constraint parameter set generation.

[0085] After the theoretical maximum flow capacity, health index, and risk level of each stroke segment have been written into the joint data table, the modeling unit uses this information to generate the safety opening upper limit and the maximum allowable action speed for the front end. First, the geometric opening ratio of the stroke segment is calculated, which is the angle range of the stroke segment divided by the total angle range of the effective stroke of the valve, to obtain the opening ratio of this segment in geometry. Then the segment flow utilization ratio is calculated, which is the target flow allowed maximum value divided by the theoretical maximum flow capacity of the stroke segment, to obtain the relative capacity required to be occupied when only considering flow demand. When the segment flow utilization ratio is less than one, theoretically no complete segment is needed to meet the target flow; when the segment flow utilization ratio is greater than one, it means that the target flow cannot be borne by a single segment and the cooperation of multiple segments is needed.

[0086] The modeling unit constructs the opening utilization coefficient by comparing the segment flow utilization ratio with the health index, taking the smaller value of the two as the opening utilization coefficient, so that even if the flow capacity of the segment is sufficient, the segment with a low health index is only allowed to participate in work with limited opening.

[0087] The safety opening upper limit angle position is determined by the following method: taking the starting angle position of the stroke segment as the reference, calculating the product of the stroke segment angle range and the opening utilization coefficient, and adding the product value to the starting angle position to obtain the safety opening upper limit angle position. For stroke segments with a high risk level, the modeling unit can add a reduction coefficient to the above opening utilization coefficient, and the reduction ratio is preset in the configuration file, so that the high-risk segment receives more stringent opening restrictions in health constraints.

[0088] The solving logic of the maximum allowable action speed is to calculate the reference action speed from the standard full stroke action time provided by the manufacturer in the actuator selection stage, and divide the entire effective angle range by the standard full stroke time to obtain the reference action speed.

[0089] During online operation, the modeling unit multiplies the reference action speed by the segment health index for each stroke segment to obtain the maximum allowable action speed. The segment with a health index close to one operates close to the reference action speed, while the segment with a lower health index has a lower action speed, thereby allowing the valve to move slowly in the stroke segment with poor mechanical conditions, reducing the risk of impact and jamming.

[0090] The modeling unit arranges the starting angle position, ending angle position, upper limit angle position of the safe opening degree, maximum allowable action speed, risk level, and health index of the stroke segment into a health constraint record, indexed by the stroke segment number, and written in sequence to the health constraint parameter set. After the health constraint parameter set is generated, the back-end platform performs consistency checks, including checking whether the upper limit angle positions of the safe opening degree of adjacent stroke segments form a continuous and monotonous safe interval, checking whether the upper limit of the safe opening degree of the high-risk segment is shrunk to the preset safety range, and checking whether the maximum allowable action speed and risk level are generated for all stroke segments. After passing the checks, the health constraint parameter set is packaged together with the unique identifier of the valve, transmitted to the corresponding front-end terminal through the communication link, and a copy is retained in the back-end storage medium for future use in working condition tracing and parameter adjustment.

[0091] Through the above step S3 processing, the segment-level characteristic parameter set and the process constraint data are completely converted into the health constraint parameter set for control execution at the stroke segment level. Each stroke segment has a safe opening degree upper limit and a maximum allowable action speed matched with the current working condition, and is accompanied by a health index and a risk level description. The front-end terminal can directly project and correct the target opening degree and action time in the process control instruction based on the health constraint parameter set in step S4, so that the actual action trajectory of the inventory valve is long-term limited in the stroke interval allowed by the working condition and relatively healthy in mechanical state, thereby reducing the risk of jamming and failure downtime.

[0092] In one embodiment, after the backend platform receives the segment-level characteristic parameter set uploaded from the field front-end terminal, the on-duty engineer views the newly added record corresponding to the valve unique identifier on the maintenance terminal in the control room. The backend platform locates the circuit to which the valve belongs in the asset ledger, and reads the upstream pressure allowable range, the downstream pressure allowable range, the upper limit of the allowable pressure difference, the target flow allowable range, and the medium characteristics of the circuit from the history library of the process control device. The backend platform writes these process constraint data and the segment-level characteristic parameter set into the same joint data structure according to the stroke segment number. The backend platform then calls the global average gradient and equivalent flow coefficient relationship saved in the offline calibration file, converts the global average gradient of each stroke segment into the equivalent flow coefficient in the joint data structure, and calculates the flow capacity of each stroke segment in combination with the process constraint data. The backend platform then generates the health index of each stroke segment and gives the risk level based on the gradient reference value and the drive reference value formed by the healthy valve sample, and then obtains the upper limit angle position of the safe opening degree and the maximum allowable action speed of each stroke segment, which constitutes the health constraint parameter set. After the on-duty engineer confirms that the health constraint parameter set matches the valve unique identifier, the backend platform issues the health constraint parameter set to the corresponding front-end terminal, and the front-end terminal maintenance interface displays that the health constraint parameter set has been updated and can be used for control instruction processing.

[0093] In steps S1 to S3, the front-end terminal has completed the small stroke trial data acquisition and stroke segment characteristic identification, and the backend platform has also generated the health constraint parameter set according to the segment-level characteristic parameter set and the process constraint data, and issued it to the front-end terminal according to the valve unique identifier. In the health constraint parameter set, the stroke segment starting angle position, the stroke segment ending angle position, the stroke segment safe opening degree upper limit angle position, the stroke segment maximum allowable action speed, and the stroke segment risk level are given for each stroke segment indexed by the stroke segment number.

[0094] The inventory valve is still issued with a target opening percentage and a desired action time by the process control device during actual operation. If the front-end terminal directly converts the target opening percentage into an angle command and drives the actuator, the valve stem may move at a high speed in the high-friction stroke segment, or cross the stroke segment safe opening degree upper limit angle position. The purpose of step S4 is to correspond the health constraint parameter set and the control instruction from the process control device one by one according to the stroke segment in the front-end terminal, to project and correct in the angle target and action speed two dimensions, and to monitor the deviation during the action. Once there is a trend of deviating from the health constraint, a safety rollback action is triggered and an abnormal event record is uploaded to the backend.

[0095] S401. Control instruction analysis and action path stroke segment set determination.

[0096] After the health constraint parameter set has been stored in the front-end terminal, whenever the process control device issues a new control instruction, the front-end first receives the target opening percentage and the expected action time. The current valve stem angle position is given by the angle sensor, which is collectively referred to as the current valve stem angle position. The valve full-closed angle position and the valve full-open angle position have been written into the front-end configuration during the debugging stage.

[0097] The control unit first calculates the target angle position according to the target opening percentage and the full-stroke angle range, which is calculated as follows: first, the full-stroke angle range is calculated, i.e., the valve full-open angle position is subtracted from the valve full-closed angle position; then, the target opening percentage is divided by one hundred to obtain a target opening coefficient; then, the full-stroke angle range is multiplied by the target opening coefficient to obtain an angle increment; finally, the angle increment is added to the valve full-closed angle position to obtain the target angle position.

[0098] The action direction is determined according to the size relationship between the current valve stem angle position and the target angle position. If the target angle position is greater than the current valve stem angle position, the action is opening; if the target angle position is less than the current valve stem angle position, the action is closing. In the health constraint parameter set, the stroke segment number corresponds to the stroke segment start angle position and the stroke segment end angle position. All stroke segment numbers are traversed one by one, and an angle interval is formed by the current valve stem angle position and the target angle position. Then, it is determined whether each stroke segment angle interval intersects with the angle interval. The specific determination method is as follows: if the stroke segment end angle position is not less than the smaller one of the current valve stem angle position and the target angle position, and the stroke segment start angle position is not greater than the larger one of the current valve stem angle position and the target angle position, it is considered that the action path will pass through this stroke segment, and the corresponding stroke segment number is added to the action path stroke segment set.

[0099] Through this process, the front-end terminal obtains a set of action path stroke segment set, which is used to constrain the angle target and the action speed in the entire action path in the subsequent process.

[0100] S402. Safe target angle position and planned angular velocity under health constraint projection are determined.

[0101] After the action path stroke segment set has been determined, the control unit needs to constrain the target angle position and the action speed by using the stroke segment safety opening upper limit angle position and the stroke segment maximum allowable action speed recorded in the health constraint parameter set. In the health constraint parameter set, each stroke segment number records the stroke segment safety opening upper limit angle position.

[0102] First, check if there is a travel segment safety opening upper limit angle position in the action path travel segment set, which is between the current valve rod angle position and the target angle position and close to the target angle position in the action direction. If the travel segment safety opening upper limit angle position of each travel segment in the action path travel segment set is not lower than the target angle position, the safety target angle position is directly taken as the target angle position. If there is one or more travel segment safety opening upper limit angle positions lower than the target angle position in the action path travel segment set, the one closest to the current valve rod angle position and located in the action direction is selected as the safety target angle position of this action. In this way, it can be ensured that the subsequent action will not exceed the travel segment safety opening upper limit angle position of any path travel segment.

[0103] After determining the safety target angle position, the control unit calculates the average command angular velocity according to the difference between the safety target angle position and the current valve rod angle position and the expected action time. The calculation process is as follows: take the absolute value of the difference between the safety target angle position and the current valve rod angle position, and divide it by the expected action time to obtain the average command angular velocity. At the same time, the control unit reads the travel segment maximum allowable action speed of each travel segment in the action path travel segment set, and takes the minimum value as the path maximum angular velocity. The determination rule of the planning angular velocity is as follows: if the average command angular velocity is not higher than the path maximum angular velocity, the planning angular velocity is equal to the average command angular velocity; if the average command angular velocity is higher than the path maximum angular velocity, the planning angular velocity is equal to the path maximum angular velocity.

[0104] Through the above two-step processing, the target angle position is projected as the safety target angle position, and the action speed is limited to the planning angular velocity that does not exceed the travel segment maximum allowable action speed of any travel segment on the action path.

[0105] S403. Planning angle position sequence construction and angle deviation amount monitoring logic.

[0106] After the safety target angle position and the planning angular velocity have been determined, the control unit needs to generate a planning angle position sequence and monitor the angle deviation amount between the actual angle position and the planning angle position in real time during execution. The front-end terminal configures a sampling period in advance, which can be set according to the angle sensor refresh frequency and the dynamic response capability of the actuator.

[0107] First, according to the angle difference between the safety target angle position and the current valve rod angle position and the planning angular velocity, the total action execution time is calculated. The calculation process is as follows: taking the absolute value of the difference between the safety target angle position and the current valve rod angle position, dividing by the planning angular velocity, the total action execution time is obtained. If the total action execution time is less than the expected action time, the total action execution time is taken as the action time this time; if the total action execution time is greater than the expected action time, the total action execution time is still taken as the action time this time, but it can be marked in the back-end statistics that this action does not meet the original time requirement. Then the total action execution time is divided by the sampling period, and the sampling number is obtained by rounding up.

[0108] In constructing the planning angle position sequence, the current valve rod angle position is taken as the first planning angle position. If the action is open, the planning angle position at each sampling time is based on the previous planning angle position plus the product of the planning angular velocity and the sampling period, until the safety target angle position or the last sampling time is reached; if the action is closed, the planning angle position at each sampling time is based on the previous planning angle position minus the product of the planning angular velocity and the sampling period, also until the safety target angle position or the last sampling time. In this way, a planning angle position sequence distributed over the total action execution time is formed.

[0109] In actual execution process, the front-end terminal collects the angle sensor output at each sampling time to form an actual angle position sequence. The control unit calculates the angle deviation at each sampling time, which is defined as the difference between the actual angle position and the corresponding planning angle position.

[0110] In online operation, the control unit checks at each sampling time whether the absolute value of the angle deviation exceeds the angle deviation threshold, and simultaneously counts the number of times that the absolute value of the angle deviation exceeds the angle deviation threshold in consecutive sampling times. The continuous overage standard number is also obtained by multiplying the maximum continuous overage number in the health valve reference action by the safety factor in the debugging stage. When the overage number in consecutive sampling times is greater than the continuous overage number standard, it is determined that the current action deviates from the health constraint, and the safety rollback action needs to be started.

[0111] S404. Safety rollback action under abnormal trigger condition and abnormal event record reporting.

[0112] When the absolute value of the angle deviation amount exceeds the angle deviation amount threshold for more than the continuous overage number standard, the control unit immediately stops the execution of the original planned angle position sequence and initiates a safety rollback action. In order to have a clear rollback target angle position in abnormal situations, the front-end terminal records a historical safety angle position after each normal action. The historical safety angle position is defined as the actual angle position at the end of an action, and the absolute value of the angle deviation amount never exceeds the angle deviation amount threshold and the driving amount does not saturate during the entire action process.

[0113] When an abnormality occurs, the control unit takes the current actual angle position as the rollback starting point and the historical safety angle position as the rollback end point to plan a one-way rollback trajectory. The rollback angle velocity selection principle of the rollback trajectory is not higher than the path maximum angle velocity in the action path travel segment set corresponding to the current action, and not higher than the current action planning angle velocity. The specific method is to compare the planning angle velocity with the path maximum angle velocity, and take the smaller value as the rollback angle velocity. The rollback trajectory uses the same sampling period as the normal action, and at each sampling time, the planned angle position is advanced along the opposite direction at the rollback angle velocity until it approaches the historical safety angle position. During the rollback process, the front-end terminal continues to collect the actual angle position and the driving amount, and once it is detected that the absolute value of the difference between the actual angle position and the historical safety angle position is below the preset tolerance, and the absolute value of the angle deviation amount is reduced to within the angle deviation amount threshold, it is considered that the rollback action is completed, and the actual angle position at this time is updated as the new historical safety angle position, providing a reference for the next abnormal rollback.

[0114] At the same time of completing the safety rollback, the front-end terminal constructs an abnormal event record.

[0115] The abnormal event record contains at least the following contents: First, the original target opening percentage, the target angle position converted from it, and the safety target angle position adopted this time; Second, the numbers of each travel segment in the action path travel segment set and the corresponding travel segment risk level; Third, the statistical basis explanation of the action planning angle velocity, the path maximum angle velocity, the angle deviation amount threshold, and the continuous overage number standard adopted at initialization; Fourth, the maximum absolute value of the angle deviation amount and the number of continuous overage sampling times during the action process; Fifth, the rollback starting angle position, the rollback ending angle position, and the rollback duration of the safety rollback action; Sixth, the abnormal trigger time and the abnormal event record generation time.

[0116] The abnormal event record is stored locally in the front-end terminal and then uploaded to the back-end platform through a communication link. The back-end platform classifies and aggregates the abnormal event in the asset management module according to the unique identifier of the valve. The health constraint parameter set and the segment-level characteristic parameter set generated in step S3 are compared to determine whether the health constraint parameter needs to be recalculated or whether the valve needs to be arranged for offline maintenance.

[0117] Through the specific implementation of step S4, the front-end terminal strictly introduces the upper limit angle position of the travel segment safety opening degree and the maximum allowable action speed in the health constraint parameter set into the calculation of the target angle position and action speed during the execution of the control instruction. The angle constraint and speed constraint are simultaneously applied to the action path across multiple travel segments. At the execution level, the angle deviation between the actual angle position and the planned angle position is monitored by the angle deviation threshold value and the continuous exceeding number of times based on the health valve statistical results. Once the action deviates from the health constraint, the safe rollback is performed according to the historical safe angle position, and the abnormal event record is formed to form a closed-loop data link between the front-end and the back-end. In this way, the segment-level characteristic parameter set obtained in steps S1 to S3 and the health constraint parameter set are implemented as specific angle trajectory and speed control rules in step S4, so that the inventory valve is truly within the safe travel segment defined by the health constraint during long-term operation.

[0118] In one embodiment, the production personnel issue a new target opening degree instruction to the valve on the process control device and set the desired action time. After receiving the instruction, the front-end terminal first converts the target opening degree to a target angle position, and then determines the travel segment set involved in the action path using the current valve rod angle position and the target angle position. The front-end terminal reads the safe opening degree upper limit angle position and the maximum allowable action speed corresponding to the travel segment set in the health constraint parameter set, projects the target angle position to a safe target angle position, and selects the most stringent maximum allowable action speed on the action path as the planned angular velocity. Subsequently, the angle trajectory is generated and the actuator is driven to move according to the trajectory. During the action process, the front-end terminal continuously acquires the valve rod angular displacement and the actuator driving amount, calculates the angle deviation between the actual angle trajectory and the angle trajectory, and determines according to the pre-established angle deviation threshold value and the continuous exceeding number of times threshold value. If the angle deviation continuously meets the abnormal triggering condition, the front-end terminal stops the original angle trajectory and performs rollback control according to the historical safe position, and generates an abnormal event record. The abnormal event record is written into the original target opening degree instruction, the safe target angle position, the travel segment set of the action path, the maximum angle deviation, the rollback start angle position, and the rollback termination angle position. The maintenance personnel in the control room can see that the back-end platform receives the abnormal event record and archives the record to the maintenance page of the valve, and the front-end terminal on site displays that the rollback is completed and updates the new historical safe position.

[0119] Specifically, the above merely describes preferred embodiments of the present application, but not intended to limit the present application.

[0120] The angle stability threshold, the driving stability threshold, the minimum angle change threshold, the maximum driving change threshold, the angle increment filtering threshold, the gradient change amount determination threshold, the index interval threshold, the dead zone angle threshold, the static friction determination threshold, the global angle merging threshold, the angle deviation amount threshold and the safety coefficient can be calibrated in advance through offline simulation test, or set as fixed values according to the field operation procedures.

[0121] In the description of the present specification, the description referring to the terms "one embodiment", "an example", "a specific example" and the like means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In the present specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0122] The preferred embodiments of the present application disclosed above are only used to help explain the present application. The preferred embodiments do not describe all the details of the present application, and the present application is not limited to the specific embodiments. Obviously, many modifications and variations can be made according to the content of the present specification. The present specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present application, so that the persons skilled in the art can well understand and utilize the present application. The present application is limited only by the claims and their full scope and equivalents.

Claims

1. A valve detection control method based on front and back ends, characterized by, The method comprises the steps of: S1: After the front-end terminal is powered on, self-checking and zero calibration are performed, the valve stem is driven to reciprocate and micro-move in a preset trial sequence within a local stroke of the valve, the angular displacement of the valve stem and the driving amount of the actuator are synchronously collected under a unified time reference, and are stored in time sequence to form small-stroke trial data; S2: The front-end terminal pre-processes the small-stroke trial data according to a unified time step, calculates the driving amount and displacement increment of adjacent sampling points, performs change point detection and piecewise fitting, obtains a plurality of stroke segments on the valve stem stroke, calculates segment-level characteristic parameters reflecting friction and hysteresis for each stroke segment, generates a segment-level characteristic parameter set, and uploads the segment-level characteristic parameter set to the back-end platform; S3: The back-end platform receives the segment-level characteristic parameter set, and correspondingly arranges the process constraint data of the valve loop in the stroke segment dimension, calculates the friction characteristics and flow capacity of each stroke segment based on a simplified working condition model, obtains the upper limit of the safe opening degree and the maximum allowable action speed of each stroke segment, forms a health constraint parameter set, and delivers the health constraint parameter set to the front-end terminal; S4: When the front-end terminal receives a target opening degree instruction, the target opening degree is converted into a target angle and the target angle is positioned on the stroke segment, the target angle and the action speed are corrected according to the health constraint parameter set to generate an angle trajectory, the angle deviation is monitored during execution, and when the angle deviation continuously exceeds the set standard, the last safe position is used for rollback control and the abnormal event is reported to the back-end platform.

2. The front-end and back-end based valve detection control method according to claim 1, wherein: Step S1 comprises: when the front-end terminal generates the preset trial sequence, the valve stem is driven to reciprocate and micro-move in a local range of the full stroke of the valve with the current working condition position as the center in a way of gradually increasing the stroke amplitude, and after collecting the angular displacement of the valve stem and the driving amount of the actuator, small-stroke trial data with driving amount and displacement change below a threshold value are removed, and small-stroke trial data used for stroke segment processing are reserved.

3. The front-end and back-end based valve detection control method according to claim 2, wherein: Step S1 further comprises: after completing the zero calibration, the front-end terminal determines the reference point of the angle sensor by using the mechanical limit of the clamping mechanism or small-amplitude reciprocating rotation, assigns a unique trial number to each trial action under the unified time reference, records the trial action direction, trial start time and trial end time, and stores the small-stroke trial data in the front-end local buffer area.

4. The front-end and back-end based valve detection control method according to claim 3, wherein: Step S2 comprises: when the front-end terminal pre-processes the small-stroke trial data, the angular displacement of the valve stem and the driving amount of the actuator are resampled according to a unified time step, high-frequency noise is weakened through digital filtering, the resampled results are subjected to amplitude normalization processing, and then the driving amount increment and the displacement increment of adjacent sampling points are calculated to construct an increment sequence for change point detection.

5. The front-end and back-end based valve detection control method according to claim 4, wherein: Step S2 further comprises: after completing the change point detection and segment fitting, the front-end terminal records the multiple stroke segments obtained by dividing the valve stem stroke in the stroke segment division table according to the stroke segment number, and writes the average slope, static friction peak value, hysteresis width and stroke segment type label corresponding to each stroke segment into the segment-level characteristic parameter set, and stores it in the stroke segment number as the index and uploads it to the back-end platform.

6. The front-back-end-based valve detection control method according to claim 5, characterized in that: Step S3 comprises: the process constraint data at least includes the upstream pressure allowable range, the downstream pressure allowable range, the upper limit of the allowable pressure difference, the target flow allowable range and the medium density of the loop to which the valve belongs, and the back-end platform writes the segment-level characteristic parameter set and the process constraint data into the same joint data structure in the stroke segment dimension, for constructing a simplified working condition model.

7. The front-back-end-based valve detection control method according to claim 6, characterized in that: Step S3 further comprises: the back-end platform calculates a health index representing the friction state and the driving force demand for each stroke segment based on the joint data structure, and divides the risk level for the stroke segment according to the pre-established threshold interval, and the health index and the risk level together with the stroke segment safe opening degree upper limit and the stroke segment maximum allowable action speed form the stroke segment record in the health constraint parameter set.

8. The front-back-end-based valve detection control method according to claim 7, characterized in that: Step S4 comprises: when the front-end terminal generates the angle trajectory according to the health constraint parameter set to correct the target angle and the action speed, it first performs safety projection on the target angle in all the stroke segments involved in the action path to obtain a safe target angle, then selects a path maximum action speed from the maximum allowable action speed of each stroke segment, and constructs a planned angle trajectory composed of an acceleration segment, a constant speed segment and a deceleration segment according to the safe target angle, the path maximum action speed and the sampling period.

9. The front-back-end-based valve detection control method according to claim 8, characterized in that: Step S4 further comprises: the front-end terminal performs multiple reference action tests on the healthy valve in advance when monitoring the angle deviation according to the sampling, and statistics the distribution of the absolute value of the angle deviation at each sampling time, determines the angle deviation threshold according to the distribution, and statistics the distribution of the number of continuous out-of-tolerance times in the healthy action, determines the continuous out-of-tolerance times standard according to the distribution, and uses the angle deviation threshold and the continuous out-of-tolerance times standard as the abnormal triggering conditions in online operation.

10. The front-back-end-based valve detection control method according to claim 9, characterized in that: The step S4 further comprises: when the angle deviation continuously exceeds the threshold value and triggers the fallback control, the front-end terminal records the actual angle position at which the last action ends and the angle deviation does not exceed the angle deviation threshold value as the last safe position, uses the last safe position as the fallback end point during the fallback control, and generates an abnormal event record after the fallback ends, the abnormal event record at least comprising the original target opening degree instruction, the safe target angle, the action path stroke segmentation set, the maximum angle deviation, the fallback starting angle position and the fallback termination angle position, and uploading to the back-end platform.

Images