Valve positioner integrating valve inner leakage self-checking, self-compensating and self-closing functions

Abstract

The invention relates to a valve positioner integrating valve inner leakage self-checking, self-compensating and self-closing functions, and belongs to the technical field of intelligent valve control. The method comprises the following steps: carrying out an opening degree adjustment test on the valve positioner, generating an opening degree-flow curve, and obtaining corresponding airtight change when the positioning scale of each valve positioner changes, so as to obtain inner leakage self-inspection data of the valve positioner; constructing an intelligent turn-off mapping model of the valve positioner, inputting the inner leakage self-inspection data of the valve positioner as detection parameters into the intelligent turn-off mapping model of the valve positioner, performing numerical simulation on leakage behaviors after the valve positioner is closed based on a fluid sealing mutual coupling mechanism, and generating a valve positioner tight-closing compensation scheme; executing the valve positioner tight closing compensation scheme, extracting leakage composition characteristics of corresponding positioning control data, and verifying a valve positioner sealing compensation period; and constructing a valve positioner turn-off self-evolution strategy, setting internal check update time, and returning internal leakage self-check data of the valve positioner after internal check update to the valve positioner intelligent turn-off mapping model based on a recursive execution feedback method to update an ideal turn-off flow baseline.

Description

Technical Field

[0001] This invention belongs to the field of intelligent valve control technology, specifically relating to a valve positioner that integrates valve internal leakage self-detection, self-compensation, and self-closing functions. Background Technology

[0002] Valves, as key actuators in industrial process control systems, are widely used in petrochemical, power, metallurgy, water treatment, and pharmaceutical industries. Their shut-off performance directly affects the safety, stability, and energy efficiency of the system. In actual operation, due to factors such as wear of the sealing surface, material aging, media erosion, and frequent changes in operating conditions, valves are prone to varying degrees of internal leakage when closed, leading to media leakage, decreased control accuracy, and even safety hazards.

[0003] Existing methods for detecting and controlling valve internal leakage mostly rely on manual inspections, periodic offline tests, or judgments based on single sensor thresholds, making it difficult to achieve real-time and precise evaluation of valve sealing performance. Furthermore, traditional valve positioners typically control only based on the target opening degree, lacking the ability to perceive and compensate for nonlinear behaviors such as flow residue and sealing hysteresis during valve closure. This results in the possibility of minute but persistent internal leakage even when a complete closure command is executed under actual operating conditions.

[0004] Furthermore, while some existing technologies introduce models or empirical parameters to correct the valve shut-off process, they mostly employ static compensation strategies. These strategies cannot be dynamically adjusted based on the valve's operating status and historical data, making it difficult to adapt to the characteristics of valve sealing performance evolving over time. When valve operating conditions change or the sealing condition deteriorates, the original compensation parameters often become invalid, requiring manual recalibration, which results in high maintenance costs and delayed response.

[0005] With the development of industrial automation and intelligence, how to achieve self-detection of valve internal leakage, intelligent compensation during shut-off, and adaptive evolution of control strategies without increasing the complexity of hardware structures has become a pressing technical problem in the field of valve control. Therefore, it is necessary to propose a valve intelligent shut-off and sealing compensation method and system that integrates valve operating data, sealing mechanisms, and intelligent algorithms to improve valve shut-off accuracy, extend service life, and ensure the safe and stable operation of industrial processes. Summary of the Invention

[0006] To address the aforementioned problems in the prior art, this invention provides a valve positioner that integrates self-detection, self-compensation, and self-sealing functions for internal leakage. The objective of this invention can be achieved through the following technical solutions: Valve positioner internal leakage self-test module: acquires the basic functional parameters of the valve positioner, performs multi-level opening adjustment tests on the valve positioner, establishes the opening-flow curve, acquires the air tightness change corresponding to the change of the valve positioner positioning scale, generates the flow benchmark under the ideal shut-off state, and maps the flow benchmark under the ideal shut-off state to the valve positioner sealing index based on the piecewise residual fitting algorithm to obtain the valve positioner internal leakage self-test data. Shutdown Mapping Compensation Module: Constructs an intelligent shut-down mapping model for the valve positioner. The internal leakage self-test data of the valve positioner is used as detection parameters and input into the intelligent shut-down mapping model. The intelligent shut-down mapping model of the valve positioner performs numerical simulation of the leakage behavior after the valve positioner is closed based on the fluid sealing mutual coupling mechanism. It outputs the valve positioner closing response time and the residual flow parameters after closing. Combined with the annotation binding method, valve positioner tight-closing compensation information is attached to the model nodes to generate a valve positioner tight-closing compensation scheme. Sealing compensation cycle verification module: Execute the valve positioner closed compensation scheme, identify the area and time period that are higher than the leakage threshold when the valve positioner is closed, extract the leakage composition characteristics of the corresponding positioning control data, and query the residual flow, hysteresis and internal leakage data in the valve positioner self-test log based on the leakage composition characteristics to verify the valve positioner sealing compensation cycle. Shutdown Evolution Feedback Update Module: Constructs a valve positioner shutdown self-evolution strategy, writes the output valve positioner shutdown response time and flow residue into the valve positioner sealing compensation cycle, sets the internal inspection update time, returns the valve positioner internal leakage self-inspection data updated by the internal inspection to the valve positioner intelligent shutdown mapping model, and updates the ideal shutdown flow baseline based on the recursive execution feedback method.

[0007] Specifically, the method for constructing the opening-flow curve is as follows: based on the real-time flow value corresponding to the valve positioner being gradually reduced from the fully open position to the fully closed position, the steady-state flow range of each positioning scale point is calculated, the flow attenuation characteristics of the steady-state flow range are extracted, and the sensitivity of the deviation between the actual measured flow and the theoretical fitted flow is enhanced to obtain the opening-flow curve.

[0008] Specifically, the segmented residual fitting algorithm divides the opening-flow curve into opening sub-intervals, including the valve positioner shut-off interval, based on the valve positioner positioning scale. Within the valve positioner shut-off sub-interval, residual calculation is performed on the measured flow data collected in the multi-level opening adjustment test. Characteristic parameters representing the degree of residual concentration and fluctuation amplitude are extracted as the shut-off residual sequence. Based on the preset sealing reference state, the flow benchmark under the ideal shut-off state is mapped to the valve positioner sealing index to obtain the valve positioner internal leakage self-test data.

[0009] Specifically, the construction method of the intelligent shut-off mapping model of the valve positioner is as follows: Based on the self-testing adaptive mapping architecture, the flow residual sequence and attenuation curve features of the opening sub-interval are extracted by combining the internal leakage self-test data of the valve positioner received by the input layer, the nonlinear sealing hysteresis characteristics of the valve positioner shut-off interval are quantified, and the hidden layer performs numerical simulation of the transient flow distribution and pressure fluctuation after the valve positioner is closed based on the fluid sealing mutual coupling mechanism. A valve positioner response residual mapping matrix is ​​established, and the intake leakage data in the matrix is ​​mapped to the valve positioner shut-off response time and flow residual to construct the intelligent shut-off mapping model of the valve positioner.

[0010] Specifically, the process of generating the valve positioner tight-closing compensation scheme is as follows: input the valve positioner internal leakage self-test data, opening-flow curve, and valve positioner closing residual sequence into the valve positioner intelligent closing mapping model; based on the fluid sealing mutual coupling mechanism, identify the opening interval and time period where the flow leakage after the valve positioner is closed is higher than the leakage threshold; synchronously fuse leakage simulation data with different sampling frequencies and dynamic response timing; and add closing speed fine-tuning, opening correction, and compensation time window parameters to the valve positioner node to generate the valve positioner tight-closing compensation scheme.

[0011] Specifically, the method for setting the valve positioner's tight-closing compensation information is as follows: based on the residual flow and response time data output by the valve positioner's intelligent shut-off mapping model, a tight-closing compensation table is constructed according to the valve positioner's opening scale and time nodes. The compensation coefficient is written into the valve positioner through the annotation binding method, and the corrected opening of the opening sub-interval that is higher than the leakage threshold is immediately adjusted, and the residual flow and sealing hysteresis are reduced in the next cycle of operation.

[0012] Specifically, the method for extracting the leakage characteristics is as follows: when the valve positioner is closed, the flow residue and closing response time in the valve positioner's shut-off interval are synchronized, the leakage behavior is decomposed into transient leakage component, stable residual component and hysteresis backflow component, and the weight ratio of each component in different opening sub-intervals is extracted by combining the opening scale mapping relationship, and the leakage characteristics are extracted.

[0013] Specifically, the verification method for the valve positioner sealing compensation cycle is as follows: a window verification cycle is set in combination with the compensated valve positioner sealing index. During the window verification cycle, residual flow and internal leakage data from multiple shutdown tests are collected. Existing leakage constitutes feature data through a distributed timed task framework. When the sealing index is consistently lower than the preset leakage threshold for multiple consecutive compensation cycles, the current valve positioner sealing compensation cycle is determined to be valid. Otherwise, the compensation parameters are recalculated and the cycle is corrected.

[0014] Specifically, the method for constructing the valve positioner shut-off self-evolution strategy is as follows: integrating historical valve positioner internal leakage self-test data and shut-off compensation schemes to establish a self-evolution hierarchy, which includes an offline training stage and an online update stage; in the offline training stage, the valve positioner shut-off response time and flow residual sequence are periodically obtained from the distributed log, and the valve positioner shut-off behavior under different opening ranges and operating conditions is fully modeled using a residual regression algorithm to generate initial self-evolution rules and threshold parameters, which are then stored in the self-test database; in the online update stage, the valve positioner operation events are monitored in real time, the airtightness change vector is calculated based on the recursive execution feedback, and the recursive feedback rules are adjusted based on the nearest neighbor principle. If the distance between the change vector and the existing rules exceeds the threshold, a new rule is created, low-activity rules are archived, similar rules are merged, and the global self-evolution parameters are updated.

[0015] Specifically, the intelligent shut-off mapping model for the valve positioner also includes a mapping index unit and a control decision unit; The mapping index unit is used to establish a multi-dimensional index relationship between the valve positioner opening sub-interval, the valve positioner internal leakage self-test data, the valve positioner closing response time and flow residue, locate the corresponding shutdown mapping node according to the valve positioner's current opening scale and operating parameters, and call the shut-off compensation parameters associated with the node to match the valve positioner's shutdown state and compensation strategy. The control decision unit is used to receive the shut-off compensation parameters output by the mapping index unit, combine the current valve positioner operation command with real-time flow feedback, and trigger the recursive execution feedback method when residual flow or shut-off deviation exceeds the threshold, to correct the shut-off compensation strategy in real time and output the final valve positioner shut-off control decision.

[0016] Specifically, after each valve positioner shut-off operation, the recursive execution feedback method compares the actual shut-off response time and residual flow with the predicted value, calculates the deviation vector as the recursive feedback input, and writes the feedback result into the self-evolution strategy to correct the valve positioner's internal leakage self-test parameters, update the shut-off compensation mapping relationship, and adjust the leakage threshold, thus completing the closed-loop execution of self-testing, compensation, and verification.

[0017] Specifically, the airtightness change vector adjusts the recursive feedback rule based on the nearest neighbor principle. If the distance between the change vector and the existing rule exceeds a threshold, a new rule is created, low-activity rules are archived, similar rules are merged, and the global self-evolution parameters are updated.

[0018] The beneficial effects of this invention are as follows: Compared with the prior art, the valve positioner proposed in this invention, which integrates valve internal leakage self-detection, self-compensation and self-closing functions, achieves refined control of the valve closing process and continuous optimization of sealing performance by introducing valve internal leakage self-detection, intelligent shutdown mapping and self-evolution control mechanism, and has the following beneficial effects.

[0019] By constructing an opening-flow curve and using a piecewise residual fitting algorithm, the ideal shut-off flow baseline is mapped to the valve sealing index, realizing a quantitative self-inspection of the valve's internal leakage status. Compared with traditional methods that rely on experience thresholds or manual detection, it can identify minute internal leaks and sealing performance degradation trends earlier and more accurately.

[0020] A valve intelligent shut-off mapping model was established based on the fluid sealing mutual coupling mechanism. The valve shut-off response time and flow residue were numerically simulated, and the shut-off process was dynamically corrected by combining a shut-off compensation scheme. This effectively reduced the residual flow and sealing hysteresis after valve shut-off, and improved the valve shut-off accuracy and operational reliability.

[0021] By introducing a self-evolutionary shutdown strategy and a recursive execution feedback mechanism, and integrating historical internal leakage self-check data and compensation results, the control parameters are continuously learned and updated online. This enables the valve shutdown strategy to adaptively adjust with changes in operating conditions and sealing status, reducing manual calibration and maintenance costs.

[0022] Without the need for additional complex hardware structures, it can be directly integrated into existing valve positioners and control systems, making it highly applicable and cost-effective in deployment. It can significantly improve the safety, stability, and energy efficiency of industrial process control systems. Attached Figure Description

[0023] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0024] Figure 1 This is a schematic diagram of the frame of a valve positioner that integrates the functions of self-detection, self-compensation and self-closing of internal leakage of valves according to the present invention.

[0025] Figure 2 This is a schematic diagram of the intelligent valve shut-off mapping model in a valve positioner that integrates valve internal leakage self-detection, self-compensation, and self-sealing functions according to the present invention. Detailed Implementation

[0026] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided.

[0027] Please see Figure 1 A valve positioner integrating internal leakage self-detection, self-compensation, and self-sealing functions: Valve positioner internal leakage self-test module: acquires the basic functional parameters of the valve positioner, performs multi-level opening adjustment tests on the valve positioner, establishes the opening-flow curve, acquires the air tightness change corresponding to the change of the valve positioner positioning scale, generates the flow benchmark under the ideal shut-off state, and maps the flow benchmark under the ideal shut-off state to the valve positioner sealing index based on the piecewise residual fitting algorithm to obtain the valve positioner internal leakage self-test data. Shutdown Mapping Compensation Module: Constructs an intelligent shut-down mapping model for the valve positioner. The internal leakage self-test data of the valve positioner is used as detection parameters and input into the intelligent shut-down mapping model. The intelligent shut-down mapping model of the valve positioner performs numerical simulation of the leakage behavior after the valve positioner is closed based on the fluid sealing mutual coupling mechanism. It outputs the valve positioner closing response time and the residual flow parameters after closing. Combined with the annotation binding method, valve positioner tight-closing compensation information is attached to the model nodes to generate a valve positioner tight-closing compensation scheme. Sealing compensation cycle verification module: Execute the valve positioner closed compensation scheme, identify the area and time period that are higher than the leakage threshold when the valve positioner is closed, extract the leakage composition characteristics of the corresponding positioning control data, and query the residual flow, hysteresis and internal leakage data in the valve positioner self-test log based on the leakage composition characteristics to verify the valve positioner sealing compensation cycle. Shutdown Evolution Feedback Update Module: Constructs a valve positioner shutdown self-evolution strategy, writes the output valve positioner shutdown response time and flow residue into the valve positioner sealing compensation cycle, sets the internal inspection update time, returns the valve positioner internal leakage self-inspection data updated by the internal inspection to the valve positioner intelligent shutdown mapping model, and updates the ideal shutdown flow baseline based on the recursive execution feedback method.

[0028] In this embodiment, the method for constructing the opening-flow curve is as follows: based on the real-time flow value corresponding to the valve positioner being gradually reduced from the fully open position to the fully closed position, the steady-state flow range of each positioning scale point is calculated, the flow attenuation characteristics of the steady-state flow range are extracted, and the sensitivity of the deviation between the actual measured flow and the theoretical fitted flow is enhanced to obtain the opening-flow curve.

[0029] In this embodiment, the piecewise residual fitting algorithm divides the opening-flow curve into opening sub-intervals, including the valve positioner shut-off interval, according to the valve positioner positioning scale. Within the valve positioner shut-off sub-interval, residual calculation is performed on the measured flow data collected in the multi-level opening adjustment test. Characteristic parameters representing the degree of residual concentration and fluctuation amplitude are extracted as the shut-off residual sequence. Based on the preset sealing reference state, the flow benchmark under the ideal shut-off state is mapped to the valve positioner sealing index to obtain the valve positioner internal leakage self-test data.

[0030] In this embodiment, a butterfly valve in a typical industrial pipeline system is used as an example. The butterfly valve is DN200, the working medium is water, and the maximum flow rate is 100 m³ / s. 3 / h, fully open 100%, fully closed 0%. The valve positioner is an intelligent pneumatic positioner equipped with a flow sensor and a pressure sensor. The test environment is normal temperature and pressure, and the initial valve sealing index is set to the ideal value of 0 (no leakage). The specific implementation process is described in detail below according to the method steps, and quantitative data is provided.

[0031] First, obtain the basic functional parameters of the valve positioner: positioning accuracy ±0.5%, response time <2s, and air source pressure 0.4-0.7MPa. Perform a multi-stage opening adjustment test on the valve, gradually decreasing the opening from the fully open position (100%) to the fully closed position (0%), collecting real-time flow values ​​at 5% opening intervals. Repeat the test three times, and take the average value as the measured flow rate.

[0032] The specific measured data are as follows (opening degree unit: %; flow rate unit: m³ / s). 3 / h): Opening size: [0, 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, 100]; Measured flow rates: [0.75, 1.18, 2.82, 5.01, 6.38, 9.13, 13.29, 16.63, 20.27, 25.52, 30.27, 36.02, 42.62, 48.29, 55.64, 63.97, 71.99, 81.41, 90.05, 99.54, 111.23].

[0033] Record the airtightness change corresponding to each opening degree: for example, in the 0-10% opening degree range, the airtightness pressure fluctuation is ±0.02MPa. The flow rate decay characteristics within the valve shut-off interval of the steady-state flow range are extracted through multivariate regression fitting. These characteristics refer to various quantitative indicators and behavioral patterns of the flow rate gradually decreasing from the initial value to a steady state (or residual value) during valve shut-off, including: decay rate, time constant, residual sequence, and backflow or delay effects during the decay process. The flow rate decay characteristics include... The fitting function adopts a quadratic form: Q = a * opening degree 2 + b * opening + c, where the fitting parameters are a=0.0101, b=0.0813, c=0.9712.

[0034] The calculation rule for the steady-state interval of the flow rate is as follows: Scan backwards from the end of the sequence to find the first point i, where |ΔQ[j]| < ΔQ_threshold for K consecutive points (K = window size), and the standard deviation σ < σ_threshold (e.g., σ_threshold = 0.02). The steady-state start time is the first time point in the sequence that meets the conditions. In this embodiment, the steady state begins 2 seconds after shutdown (rate of change < 0.01 m). 3 ( / h / s), use multinomial regression fitting (e.g., quadratic function) to check for nonlinear deviation in the steady-state interval. If the mean and standard deviation of the residual sequence are below the threshold (e.g., mean <0.1, standard deviation <0.05), the interval is confirmed to be valid.

[0035] Sensitivity enhancement is applied to the deviation between the actual measured flow rate and the theoretically fitted flow rate to obtain the opening-flow curve. The nonlinear change in valve sealing performance within the shut-off interval (0-10%) is represented by a residual sequence: [-0.223, -0.450, 0.028]. Based on a piecewise residual fitting algorithm, the opening-flow curve is divided into sub-intervals (e.g., 0-10%, 10-50%, 50-100%). Within the shut-off sub-intervals, residuals are calculated on the measured flow data, and the residual concentration (mean -0.215) and fluctuation amplitude (standard deviation 0.195) are extracted as the shut-off residual sequence. The preset sealing reference state is residual <0.1, and the ideal shut-off flow baseline (0 m) is used. 3 / h) is mapped to the valve sealing index -1.102 (a negative value indicates a slight tendency to leak), yielding the valve internal leakage self-test data: internal leakage 0.5 m 3 / h, hysteresis error 1.2%.

[0036] Construct a valve intelligent shut-off mapping model and generate a valve tight-closing compensation scheme. Construct a valve intelligent shut-off mapping model: Based on the self-testing adaptive mapping architecture, the input layer receives valve internal leakage self-test data (sealing index -1.102, residual sequence [-0.223, -0.450, 0.028]), and extracts the flow residual sequence and attenuation curve features of the opening sub-interval (the nonlinear sealing hysteresis characteristic is quantized to 0.8).

[0037] The hidden layer is based on the fluid-sealed coupling mechanism (considering the coupling between fluid viscosity and valve disc deformation) to numerically simulate the leakage behavior after valve closure. A simplified ODE model is used to simulate transient flow distribution and pressure fluctuations: the flow decay equation is dQ / dt = -Q / τ + residual / τ, where the time constant τ = 1.0 s and the initial flow rate is 10 m³ / s. 3 / h, residual flow rate 0.5m 3 / h.

[0038] Simulation results: Time series (s): [0, 0.1, 0.2, ..., 10]; Flow series (first 10 points): [10.00, 9.10, 8.28, 7.54, 6.87, 6.26, 5.71, 5.22, 4.77, 4.36]; Output valve closing response time: 10.0 s; Residual flow parameter after closing: 0.50 m³ / s. 3 / h.

[0039] Establish a valve response residual mapping matrix (example: rows represent opening sub-intervals, columns represent time nodes), and map leakage data to the matrix elements. Combined with annotation binding, attach valve tight-closing compensation information to the model nodes: fine-tune closing speed +0.5s, correct opening degree -0.5%, compensation time window 3s. Generate a valve tight-closing compensation scheme: for values ​​above the leakage threshold (0.1 m... 3 The valve opening range (0-5%) is simultaneously fused with leakage simulation data with a sampling frequency of 10Hz to construct a tight-closing compensation mapping table, and a compensation coefficient of 0.95 is written into the valve positioner.

[0040] Implement the valve tightness compensation scheme and verify the valve sealing compensation cycle. Implement compensation measures: Identify valves with leakage exceeding the threshold by 0.1 m when the valve is closed. 3 / h area (opening degree 0-5%) and time period (0-2s after closing). Extract the leakage characteristics of the corresponding positioning control data: in the closed state, the residual flow in the closed interval is 0.50 m. 3 / h and the shutdown response time of 10.0s are synchronously decomposed into a transient leakage component (0.30 m). 3 / h, weighted at 60%), stable residual component (0.15 m) 3 / h, weight 30%) and hysteresis backflow component (0.05 m) 3 / h, weight 10%).

[0041] Based on the leakage characteristics, the valve flow profile (from opening degree to flow rate curve) and the residual flow rate of 0.50 m³ in the self-test log were queried. 3 / h, hysteresis 1.2% and internal leakage 0.5 m 3 / h data. Verification of valve sealing compensation cycle: After compensation, the sealing index changed from -1.102 to -0.50, and the calculated compensation effectiveness was 54.5%. A 24-hour window verification cycle was set, and data from 5 shut-off tests were collected within the window (average residual flow rate 0.45 m³ / h). 3 / h, internal leakage 0.48 m 3 / h). By reusing leakage characteristics through a distributed timed task framework, when key sealing indicators (residual <0.3 m) are maintained for three consecutive cycles... 3When the value of / h is consistently below the threshold, the compensation cycle is deemed valid; otherwise, a recalculation is triggered.

[0042] Construct a valve shut-off self-evolution strategy and update the ideal shut-off flow baseline. Constructing a valve shut-off self-evolution strategy: Integrating historical internal leakage self-check data (sealing index sequence [-1.102, -0.50]) and a tight-closing compensation scheme, a self-evolutionary hierarchy is established, including an offline training phase (periodically obtaining shut-off response time of 10.0s and flow residue of 0.50 m from logs). 3 The / h sequence is modeled using a residual regression algorithm to generate initial rules: a threshold of 0.2 m. 3 / h) and online update phase (real-time event monitoring, calculation of air tightness change vector [0.05, -0.02] based on recursive execution feedback, adjustment of rules, creation of new rules if distance > 0.1, and merging of similar rules).

[0043] The output shutdown response time of 10.0s and the residual flow of 0.50 m³ are set. 3 The valve sealing compensation cycle is written in / h, and the internal inspection update time is set to 12h. Based on the recursive execution feedback method, the internal leakage self-inspection data after the internal inspection update (new sealing index -0.45) is returned to the valve intelligent shutdown mapping model, updating the ideal shutdown flow baseline from 0 m³. 3 / h to 0.05 m 3 / h.

[0044] Establish a multi-dimensional index: the opening sub-interval of 0-10% corresponds to the compensation parameter 0.95 and the control decision unit (receives parameters, triggers the recursive feedback correction strategy, and outputs the final decision: opening correction -0.3%).

[0045] Recursive execution feedback method: After each shutdown, the actual response time of 9.8s is compared with the residual time of 0.48m. 3 The error vector [0.2, 0.02] is calculated by comparing / h with the predicted value and used as input to correct the internal leakage self-check parameters (updating the threshold to 0.15 m). 3 / h), close the compensation mapping relationship, and adjust the leakage threshold to 0.08 m. 3 / h completes the closed-loop execution.

[0046] In this embodiment, the method for constructing the intelligent shut-off mapping model of the valve positioner is as follows: Based on the self-testing adaptive mapping architecture, the flow residual sequence and attenuation curve features of the opening sub-interval are extracted by combining the self-testing data of the valve positioner's internal leakage received by the input layer, the nonlinear sealing hysteresis characteristics of the valve positioner's shut-off interval are quantified, and the hidden layer performs numerical simulation of the transient flow distribution and pressure fluctuation after the valve positioner is closed based on the fluid sealing mutual coupling mechanism. A valve positioner response residual mapping matrix is ​​established, and the intake leakage data in the matrix is ​​mapped to the valve positioner's shut-off response time and flow residual to construct the intelligent shut-off mapping model of the valve positioner.

[0047] In this embodiment, the process of generating the valve positioner tight-closing compensation scheme is as follows: the valve positioner internal leakage self-test data, opening-flow curve and valve positioner closing residual sequence are input into the valve positioner intelligent closing mapping model. Based on the fluid sealing mutual coupling mechanism, the opening interval and time period where the flow leakage after the valve positioner is closed is identified as higher than the leakage threshold. The leakage simulation data with different sampling frequencies and dynamic response timing are synchronously fused. The closing speed fine-tuning, opening correction and compensation time window parameters are added to the valve positioner node to generate the valve positioner tight-closing compensation scheme.

[0048] In this embodiment, a pneumatic control valve used in an industrial pipeline system is taken as the object. This control valve is equipped with an intelligent valve positioner, which has valve position feedback, stroke control, and flow signal acquisition functions. First, the basic functional parameters of the valve positioner are acquired. These basic functional parameters include the valve's full stroke range, positioning scale resolution, maximum and minimum valve disc displacement values, actuator response time, valve position feedback accuracy, and flow sensor sampling period, etc., which are used to characterize the valve's basic control capability and detection accuracy. An opening adjustment test is then performed on the valve. Specifically, under stable operating conditions, the valve is controlled to gradually decrease its opening from a fully open state according to a preset positioning scale until it is fully closed. A set time is maintained at each opening scale to obtain a stable flow value, and the valve displacement signal and corresponding real-time flow data are recorded simultaneously. By filtering and fitting the acquired data, a valve opening-flow curve is generated, and the airtightness changes corresponding to each positioning scale are recorded.

[0049] The piecewise residual fitting algorithm is used to analyze the opening-flow curve. The shut-off interval is divided into multiple opening sub-intervals. The residuals of the measured flow data collected in the reciprocating opening and closing test and the ideal shut-off flow baseline are calculated. The residual characteristics are mapped to the valve sealing index, thereby forming valve internal leakage self-test data.

[0050] The ideal shut-off flow baseline Q0 is mapped to the valve sealing index SI by the residual characteristic. i : , Where λ is the residual fluctuation weighting coefficient, and ε is the stability factor to prevent the denominator from being zero.

[0051] A valve response residual mapping matrix is ​​constructed, with the valve opening sub-interval and operating parameters as index dimensions. Matrix elements correspond to the valve's response time and flow residual value during the closing process. By introducing a fluid-sealing mutual coupling mechanism, the transient flow distribution and sealing hysteresis behavior after valve closure are numerically simulated. The simulation results are fused with measured data to fill and update the valve response residual mapping matrix. This matrix serves as the core data structure of the valve intelligent shut-off mapping model, used to generate shut-off compensation parameters. During subsequent operation, the matrix content is dynamically corrected using a recursive execution feedback method, achieving continuous optimization of valve shut-off performance.

[0052] The mapping matrix is ​​used to construct the valve response residual mapping matrix. The residual flow rate after closure is defined as follows: , , Where the row index is the open sub-interval i, the column index is the working condition number k, and the element is a binary tuple M. i，k =(t i，k q i，k ), t close This refers to the moment when the traffic drops to the threshold.

[0053] In this embodiment, the method for setting the valve positioner tight-closing compensation information is as follows: based on the residual flow and response time data output by the valve positioner intelligent shut-off mapping model, a tight-closing compensation table is constructed according to the valve positioner opening scale and time nodes. The compensation coefficient is written into the valve positioner through the annotation binding method, and the correction opening of the opening sub-interval higher than the leakage threshold is immediately adjusted, and the residual flow and sealing hysteresis are reduced in the next cycle of operation.

[0054] In this embodiment, a distributed timed task framework is used to schedule and manage the valve control system. This framework includes a task scheduling unit, a node execution unit, a data synchronization unit, and a status monitoring unit. The task scheduling unit assigns tasks such as valve internal leakage self-inspection, opening-flow curve acquisition, and valve shut-off residual analysis to each control node according to a preset cycle or event triggering conditions. The node execution unit performs tasks locally, including collecting data such as valve opening, inlet and outlet pressure, flow rate and air tightness changes, calculating shut-off residuals and generating local tightness compensation parameters; The data synchronization unit transmits the calculation results of each node back to the centralized or distributed database in real time to achieve global state synchronization and data consistency. The status monitoring unit monitors the task execution status of each node in real time, and provides task retry, dynamic load balancing and exception handling mechanisms to ensure the high reliability and low latency execution of the valve shutdown compensation strategy in a multi-node environment.

[0055] A closing-tightness compensation table is constructed for each valve to describe the mapping relationship between leakage characteristics at each stage of the valve closing process and the corresponding compensation control quantities. The closing-tightness compensation table is indexed by opening range, time period, and valve response parameters, and includes closing speed fine-tuning, target opening correction, and compensation time window parameters as table entries.

[0056] In this embodiment, the method for extracting the leakage characteristics is as follows: when the valve positioner is closed, the flow residue and closing response time in the valve positioner's closing interval are synchronized, the leakage behavior is decomposed into transient leakage component, stable residual component and hysteresis backflow component, and the weight ratio of each component in different opening sub-intervals is extracted by combining the opening scale mapping relationship, and the leakage characteristics are extracted.

[0057] In this embodiment, the verification method for the valve positioner sealing compensation cycle is as follows: a window verification cycle is set in combination with the compensated valve positioner sealing index. During the window verification cycle, residual flow and internal leakage data from multiple shutdown tests are collected. Existing leakage constitutes feature data through a distributed timed task framework. When the sealing index is consistently lower than the preset leakage threshold for multiple consecutive compensation cycles, the current valve positioner sealing compensation cycle is determined to be valid. Otherwise, the compensation parameters are recalculated and the cycle is corrected.

[0058] In this embodiment, the method for constructing the valve positioner shut-off self-evolution strategy is as follows: integrating historical valve positioner internal leakage self-test data and shut-off compensation schemes to establish a self-evolution hierarchy, which includes an offline training stage and an online update stage; in the offline training stage, the valve positioner shut-off response time and flow residual sequence are periodically obtained from the distributed log, and the valve positioner shut-off behavior under different opening ranges and operating conditions is fully modeled using a residual regression algorithm to generate initial self-evolution rules and threshold parameters, which are then stored in the self-test database; in the online update stage, the valve positioner operation events are monitored in real time, the airtightness change vector is calculated based on the recursive execution feedback, and the recursive feedback rules are adjusted based on the nearest neighbor principle. If the distance between the change vector and the existing rules exceeds the threshold, a new rule is created, low-activity rules are archived, similar rules are merged, and the global self-evolution parameters are updated.

[0059] In this embodiment, a smart valve shut-off mapping model is constructed for valve shut-off process modeling and compensation decision-making. This model characterizes the valve's shut-off response behavior and residual flow characteristics under different opening ranges and operating conditions. Based on the valve response residual mapping matrix, a label binding method is used to generate shut-off compensation parameters on the mapping matrix nodes. , Where K is the compensation gain matrix, and each mapping node forms a one-to-one binding relationship with its corresponding compensation parameter, avoiding overfitting or failure problems caused by global uniform compensation.

[0060] After the valve is actually in operation, the measured values ​​are obtained. , And update the mapping model parameters using an exponential recursive approach: , Where η is the learning rate, the mapping model can achieve online adaptive updates without retraining the entire model.

[0061] In this embodiment, as Figure 2 The intelligent shut-off mapping model for the valve positioner shown also includes a mapping index unit and a control decision unit; The mapping index unit is used to establish a multi-dimensional index relationship between the valve positioner opening sub-interval, the valve positioner internal leakage self-test data, the valve positioner closing response time and flow residue, locate the corresponding shutdown mapping node according to the valve positioner's current opening scale and operating parameters, and call the shut-off compensation parameters associated with the node to match the valve positioner's shutdown state and compensation strategy. The control decision unit is used to receive the shut-off compensation parameters output by the mapping index unit, combine the current valve positioner operation command with real-time flow feedback, and trigger the recursive execution feedback method when residual flow or shut-off deviation exceeds the threshold, to correct the shut-off compensation strategy in real time and output the final valve positioner shut-off control decision.

[0062] In this embodiment, after each valve positioner shut-off operation, the recursive execution feedback method compares the actual shut-off response time and residual flow with the predicted value, calculates the deviation vector as the recursive feedback input, and writes the feedback result into the self-evolution strategy to correct the valve positioner's internal leakage self-test parameters, update the shut-off compensation mapping relationship, and adjust the leakage threshold, thus completing the closed-loop execution of self-testing, compensation, and verification.

[0063] In this embodiment of the invention, the airtightness change vector adjusts the recursive feedback rule based on the nearest neighbor principle. If the distance between the change vector and the existing rule exceeds a threshold, a new rule is created, low-activity rules are archived, similar rules are merged, and the global self-evolution parameters are updated.

[0064] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A valve positioner integrating self-detection, self-compensation, and self-sealing functions for internal leakage of valves, characterized in that, include: Valve positioner internal leakage self-test module, shut-off mapping compensation module, sealing compensation cycle verification module, shut-off evolution feedback update module; Valve positioner internal leakage self-test module: acquires the basic functional parameters of the valve positioner, performs multi-level opening adjustment tests on the valve positioner, establishes the opening-flow curve, acquires the air tightness change corresponding to the change of the valve positioner positioning scale, generates the flow benchmark under the ideal shut-off state, and maps the flow benchmark under the ideal shut-off state to the valve positioner sealing index based on the piecewise residual fitting algorithm to obtain the valve positioner internal leakage self-test data. Shutdown Mapping Compensation Module: Constructs an intelligent shut-down mapping model for the valve positioner. The internal leakage self-test data of the valve positioner is used as detection parameters and input into the intelligent shut-down mapping model. The intelligent shut-down mapping model of the valve positioner performs numerical simulation of the leakage behavior after the valve positioner is closed based on the fluid sealing mutual coupling mechanism. It outputs the valve positioner closing response time and the residual flow parameters after closing. Combined with the annotation binding method, valve positioner tight-closing compensation information is attached to the model nodes to generate a valve positioner tight-closing compensation scheme. Sealing compensation cycle verification module: Execute the valve positioner closed compensation scheme, identify the area and time period that are higher than the leakage threshold when the valve positioner is closed, extract the leakage composition characteristics of the corresponding positioning control data, and query the residual flow, hysteresis and internal leakage data in the valve positioner self-test log based on the leakage composition characteristics to verify the valve positioner sealing compensation cycle. Shutdown Evolution Feedback Update Module: Constructs a valve positioner shutdown self-evolution strategy, writes the output valve positioner shutdown response time and flow residue into the valve positioner sealing compensation cycle, sets the internal inspection update time, returns the valve positioner internal leakage self-inspection data updated by the internal inspection to the valve positioner intelligent shutdown mapping model, and updates the ideal shutdown flow baseline based on the recursive execution feedback method.

2. The valve positioner according to claim 1, characterized in that, The method for constructing the opening-flow curve is as follows: based on the real-time flow value corresponding to the valve positioner being gradually reduced from the fully open position to the fully closed position, the steady-state flow range of each positioning scale point is calculated, the flow attenuation characteristics of the steady-state flow range are extracted, and the sensitivity of the deviation between the actual measured flow and the theoretical fitted flow is enhanced to obtain the opening-flow curve.

3. The valve positioner according to claim 1, characterized in that, The segmented residual fitting algorithm divides the opening-flow curve into opening sub-intervals, including the valve positioner shut-off interval, based on the valve positioner positioning scale. Within the valve positioner shut-off sub-interval, residual calculation is performed on the measured flow data collected in the multi-level opening adjustment test. Characteristic parameters representing the degree of residual concentration and fluctuation amplitude are extracted as the shut-off residual sequence. Based on the preset sealing reference state, the flow benchmark under the ideal shut-off state is mapped to the valve positioner sealing index to obtain the valve positioner internal leakage self-test data.

4. The valve positioner according to claim 1, characterized in that, The method for constructing the intelligent shut-off mapping model of the valve positioner is as follows: Based on the self-testing adaptive mapping architecture, the flow residual sequence and attenuation curve features of the opening sub-interval are extracted by combining the self-testing data of the valve positioner's internal leakage received by the input layer. The nonlinear sealing hysteresis characteristics of the valve positioner's shut-off interval are quantified. The hidden layer performs numerical simulation of the transient flow distribution and pressure fluctuation after the valve positioner is closed based on the fluid sealing mutual coupling mechanism. A valve positioner response residual mapping matrix is ​​established, and the intake leakage data in the matrix is ​​mapped to the valve positioner's shut-off response time and flow residual to construct the intelligent shut-off mapping model of the valve positioner.

5. The valve positioner according to claim 2, characterized in that, The process of generating the valve positioner tight-closing compensation scheme is as follows: input the valve positioner internal leakage self-test data, opening-flow curve and valve positioner closing residual sequence into the valve positioner intelligent closing mapping model; based on the fluid sealing mutual coupling mechanism, identify the opening interval and time period where the flow leakage after the valve positioner is closed is higher than the leakage threshold; synchronously fuse the leakage simulation data with different sampling frequencies and dynamic response timing; and add closing speed fine-tuning, opening correction and compensation time window parameters to the valve positioner positioner node to generate the valve positioner tight-closing compensation scheme.

6. The valve positioner according to claim 5, characterized in that, The method for setting the valve positioner's tight-closing compensation information is as follows: based on the residual flow and response time data output by the valve positioner's intelligent shut-off mapping model, a tight-closing compensation table is constructed according to the valve positioner's opening scale and time nodes. The compensation coefficient is written into the valve positioner through the annotation binding method, and the corrected opening degree of the opening sub-interval that is higher than the leakage threshold is immediately adjusted, and the residual flow and sealing hysteresis are reduced in the next cycle of operation.

7. The valve positioner according to claim 4, characterized in that, The method for extracting the leakage characteristics is as follows: when the valve positioner is closed, the flow residue and closing response time in the valve positioner's closing interval are synchronized, the leakage behavior is decomposed into transient leakage component, stable residual component and hysteresis backflow component, and the weight ratio of each component in different opening sub-intervals is extracted by combining the opening scale mapping relationship, and the leakage characteristics are extracted.

8. The valve positioner according to claim 2, characterized in that, The verification method for the valve positioner sealing compensation cycle is as follows: a window verification cycle is set in combination with the valve positioner sealing index after compensation. During the window verification cycle, residual flow and internal leakage data from multiple shutdown tests are collected. Existing leakage constitutes feature data through a distributed timed task framework. When the sealing index is consistently lower than the preset leakage threshold for multiple consecutive compensation cycles, the current valve positioner sealing compensation cycle is determined to be valid. Otherwise, the compensation parameters are recalculated and the cycle is corrected.

9. The valve positioner according to claim 4, characterized in that, The method for constructing the valve positioner shut-off self-evolution strategy is as follows: integrating historical valve positioner internal leakage self-test data and shut-off compensation schemes to establish a self-evolution hierarchy, which includes an offline training stage and an online update stage; in the offline training stage, the valve positioner shut-off response time and flow residual sequence are periodically obtained from the distributed log, and the valve positioner shut-off behavior under different opening ranges and operating conditions is fully modeled using a residual regression algorithm to generate initial self-evolution rules and threshold parameters, which are then stored in the self-test database; The line update phase monitors valve positioner operation events in real time, calculates the airtightness change vector based on recursive execution feedback, and adjusts the recursive feedback rules based on the nearest neighbor principle. If the distance between the change vector and the existing rules exceeds a threshold, a new rule is created, low-activity rules are archived, similar rules are merged, and the global self-evolution parameters are updated.

10. The valve positioner according to claim 1, characterized in that, The intelligent shut-off mapping model for the valve positioner also includes a mapping index unit and a control decision unit. The mapping index unit is used to establish a multi-dimensional index relationship between the valve positioner opening sub-interval, the valve positioner internal leakage self-test data, the valve positioner closing response time and flow residue, locate the corresponding shutdown mapping node according to the valve positioner's current opening scale and operating parameters, and call the shut-off compensation parameters associated with the node to match the valve positioner's shutdown state and compensation strategy. The control decision unit is used to receive the shut-off compensation parameters output by the mapping index unit, combine the current valve positioner operation command with real-time flow feedback, and trigger the recursive execution feedback method when residual flow or shut-off deviation exceeds the threshold, to correct the shut-off compensation strategy in real time and output the final valve positioner shut-off control decision.

11. The valve positioner according to claim 10, characterized in that, The recursive execution feedback method compares the actual closing response time and residual flow with the predicted value after each valve positioner shut-off operation, calculates the deviation vector as the recursive feedback input, and writes the feedback result into the self-evolution strategy to correct the valve positioner's internal leakage self-test parameters, update the tight-closing compensation mapping relationship, and adjust the leakage threshold, thus completing the closed-loop execution of self-testing, compensation, and verification.

12. The valve positioner according to claim 9, characterized in that, The airtightness change vector adjusts the recursive feedback rule based on the nearest neighbor principle. If the distance between the change vector and the existing rule exceeds a threshold, a new rule is created, low-activity rules are archived, similar rules are merged, and the global self-evolution parameters are updated.

Images