Method for diagnosing internal leakage of high-pressure slurry valve

Abstract

The invention relates to the technical field of industrial equipment operation state sensing and safety monitoring, in particular to a high-pressure slurry valve inner leakage diagnosis method. Aiming at the problem that the inner leakage state of the high-pressure slurry valve under the complex working condition is difficult to identify in an early stage, the method comprises the following steps: applying modulation disturbance to the operation working condition of the valve to obtain recovery behavior data after disturbance is relieved, and constructing a sealing self-adaptive reference behavior model; extracting a recovery behavior parameter set representing the valve sealing adjustment behavior, and obtaining the behavior offset of the current operation state of the valve relative to the reference state based on a behavior consistency analysis method; whether the high-pressure slurry valve is in an internal leakage associated state or not is judged by analyzing behavior offset evolution characteristics; the method does not depend on additional detection hardware or instantaneous abnormal signals, the evolution characteristics of the valve sealing self-adaptive capacity are reflected, prospective recognition of the high-pressure slurry valve inner leakage associated state is achieved, and the reliability and engineering applicability of slurry valve operation monitoring are improved.

Description

Technical Field

[0001] This invention relates to the field of industrial equipment operation status perception and safety monitoring technology, and more specifically to a method for diagnosing internal leakage in a high-pressure slurry valve. Background Technology

[0002] High-pressure slurry valves are widely used in high-pressure applications such as mineral slurry transportation, chemical reactions, and energy. Internal leakage faults in these valves are often difficult to detect and can easily lead to media loss and equipment damage, necessitating reliable online diagnostic methods. Existing valve internal leakage diagnostic methods primarily rely on the analysis and identification of observable physical signals during valve operation. By establishing a correspondence between signal characteristics and the valve's internal leakage status, internal leakage can be detected and assessed. Related technologies typically collect acoustic, vibration, and temperature signals from valves or pipelines, as well as operating parameters such as pressure and flow rate. Key indicators reflecting leakage characteristics are then extracted through signal processing and model analysis.

[0003] For example, CN116222905A uses acoustic signal sensors placed upstream and downstream of the valve to collect acoustic signals in real time during valve operation, and uses an algorithm model to identify the valve's internal leakage status, achieving online detection and quantitative analysis of valve internal leakage. Another example is CN118817189B, which combines acoustic signals, fluid characteristic data, and valve flaw detection data, and analyzes valve internal leakage characteristics based on a joint detection algorithm to generate valve internal leakage alarm information.

[0004] The methods described above, through analysis and modeling of multi-source signals, can identify valve internal leakage under certain operating conditions. However, their technical foundation generally rests on the premise that the valve internal leakage process can form relatively stable and sustainable characterization features at the acoustic, vibration, or thermal signal levels. Existing methods, in the process of model construction and feature extraction, typically assume that the energy release or parameter changes corresponding to the internal leakage behavior have statistical stability, thus enabling identification through threshold judgment, pattern matching, or similarity calculation.

[0005] However, in the actual operating environment of high-pressure slurry valves, the medium is usually a high-concentration, multiphase slurry. Internal leakage is often accompanied by repeated occurrences of particle embedding, localized blockage, and instantaneous scouring, resulting in significant discontinuity and strong time-varying characteristics in the leakage channels. Under these conditions, the acoustic, vibration, or temperature changes caused by internal leakage are difficult to maintain consistency over time, and the related characteristics exhibit highly discrete and non-repeatable distributions in a statistical sense. Since existing diagnostic methods rely on the extraction and learning of stable feature patterns, when the internal leakage behavior itself lacks continuous and cumulative signal manifestations, the diagnostic model struggles to form effective judgment criteria, making it difficult to accurately identify the internal leakage state in its early stages. This, in turn, affects the judgment and evaluation of the actual operating state of the high-pressure slurry valve. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention discloses a method for diagnosing internal leakage in high-pressure slurry valves, aiming to achieve proactive identification of the associated states of internal leakage in high-pressure slurry valves and improve the reliability and engineering applicability of slurry valve operation monitoring.

[0007] To achieve the above-mentioned technical effects, the present invention adopts the following technical solution: A method for diagnosing internal leakage in a high-pressure slurry valve includes the following steps: Step 1: Apply an operational modulation disturbance to the high-pressure slurry valve in the state of no internal leakage confirmation using an operational condition modulation method to induce the redistribution of medium particles at the valve sealing interface. Collect system operating parameters before and after the operational disturbance to obtain first recovery behavior data representing the process of the valve entering a stable state after the disturbance is removed. Construct a sealing adaptive reference behavior model based on the first recovery behavior data. Step 2: Apply the same or equivalent operating condition modulation method as in Step 1 to the high-pressure slurry valve in online operation, and collect the system operating parameters after the operating disturbance is removed to obtain the second recovery behavior data of the valve in the current operating state. Step 3: Perform time-scale unified processing on the first recovery behavior data and the second recovery behavior data, and extract a set of recovery behavior parameters based on the processed data to reflect the recovery path characteristics of the valve after disturbance. The set of recovery behavior parameters is used to represent the adjustment behavior of the valve sealing structure to the current operating disturbance. Step 4: Use the behavior consistency analysis method to compare the set of recovery behavior parameters with the sealing adaptive reference behavior model to obtain the behavior offset of the valve's current operating state relative to the reference state; Step 5: Determine the sealing adaptive state of the high-pressure slurry valve based on the change characteristics of the behavior offset. When the behavior offset meets the preset evolution conditions, determine that the high-pressure slurry valve is in an internal leakage associated state.

[0008] Furthermore, the operating condition modulation method includes applying periodic micro-displacement control to the valve core of the high-pressure slurry valve relative to the valve seat. The micro-displacement control range is within the displacement interval that does not change the valve's on / off state, and a preset holding time is maintained after each micro-displacement control, inducing the slurry particles at the valve seat sealing surface to rearrange their positions in the contact area.

[0009] Furthermore, the first recovery behavior data consists of a sequence of pressure changes before the valve, a sequence of pressure changes after the valve is de-energized, and a sequence of valve position feedback changes continuously collected after the operation modulation disturbance is removed. The first recovery behavior data uses the disturbance removal time as a unified time reference point and is divided into at least one unsteady-state recovery segment and one steady-state approaching segment according to the evolution characteristics of the system operating parameters, in order to construct a sealing adaptive reference behavior model to describe the staged recovery behavior of the valve sealing interface during particle redistribution. The sealing adaptive reference behavior model includes stage sequence relationship constraints to limit the order of occurrence of each recovery segment, stage connection constraints to limit the evolution continuity of adjacent recovery segments, and stability constraints to limit the fluctuation range of system parameters within the steady-state approaching segment.

[0010] Furthermore, the acquisition of the second recovery behavior data includes: When the high-pressure slurry valve is in continuous feeding operation, the valve's current operating stability is judged based on the valve position feedback signal. When the valve position feedback change is within the preset allowable range, the operating condition modulation corresponding to step one is triggered. After the modulation of the operating condition is released, the upstream pressure, downstream pressure and valve position feedback signal of the valve are collected synchronously. The collected data is segmented and the data segments before the modulation of the operating condition is removed and the data segments whose parameters show non-monotonic changes after the disturbance is removed are removed. Data segments that satisfy the monotonic recovery characteristics and whose duration exceeds the preset minimum observation time are constructed as the second recovery behavior data of the valve under the current operating state.

[0011] Furthermore, the time-scale unification processing of the first recovery behavior data and the second recovery behavior data includes: The recovery process of the valve from the moment the operational disturbance is removed to the time interval for determining the stable state is determined in the first recovery behavior data and the second recovery behavior data, respectively; Select at least one of the valve upstream pressure data, valve downstream pressure data, and valve position feedback data as a characterization of the recovery process; Based on the rate of change of the recovery process characteristic quantities during the recovery process, the physical time axis is nonlinearly reconstructed to obtain the equivalent recovery process time axis. The reconstruction expression is: in, Indicates the time when the operational disturbance is resolved; This indicates the moment when the system operating parameters enter the stable state determination interval; Represents any physical time sampling point during the recovery process; Represents the integral variable; Indicates at time The collected system operating parameters are selected from at least one of valve upstream pressure, downstream pressure, or valve position feedback. This represents the magnitude of the instantaneous rate of change of the system's operating parameters during the recovery process; This represents the normalized equivalent recovery process time variable, with a value range of [value missing]. This is used to characterize the evolution of valve disturbance recovery at the level of sealing regulation behavior; On the equivalent recovery process timeline, the first recovery behavior data and the second recovery behavior data are uniformly resampled; Based on the resampled recovery path, a set of recovery behavior parameters is extracted from the second recovery behavior data. The set of recovery behavior parameters includes at least the recovery process slope distribution parameter, the recovery path curvature parameter, and the stable region precursor evolution length parameter.

[0012] Furthermore, the process of comparing the second set of recovery behavior parameters with the sealing adaptive reference behavior model using the behavior consistency analysis method includes: Based on the sealing adaptive reference behavior model, each recovery behavior parameter in the recovery behavior parameter set is numbered according to its order of appearance on the equivalent recovery process time axis to construct a reference recovery behavior sequence; The direction of change, range of change, and evolution interval of adjacent recovery behavior parameters in the reference recovery behavior sequence are statistically analyzed to generate a parameter correlation matrix that characterizes the intrinsic constraint relationship of recovery behavior under the reference state. Based on the set of recovery behavior parameters, the actual recovery behavior sequence under the current running state is constructed according to the same numbering rule, and the corresponding actual parameter association matrix is ​​generated; The actual parameter correlation matrix is ​​compared with the reference parameter correlation matrix item by item to identify correlation items that violate the constraints of change direction, amplitude range, or evolution interval. Based on the distribution density and deviation magnitude of the violation associated terms on the equivalent recovery process time axis, the behavioral offset of the valve's current operating state relative to the reference state is quantitatively characterized.

[0013] Furthermore, the construction of the parameter correlation matrix includes: On the equivalent recovery process timeline, for any two adjacent recovery behavior parameters and The monotonic change direction identifier, cumulative change amplitude ratio, and equivalent process interval are calculated respectively within the recovery process interval; wherein, the monotonic change direction identifier is determined by the first-order change sign of the recovery behavior parameter within the corresponding recovery process interval; the formula for calculating the equivalent process interval is... The formula for calculating the cumulative change amplitude ratio is as follows: in, They represent the first The and the first One recovery behavior parameter, This refers to the position of the corresponding parameter on the equivalent recovery process timeline. The cumulative change magnitude ratio of adjacent recovery behavior parameters; The monotonic change direction identifier, cumulative change amplitude ratio, and equivalent process interval are used as correlation elements to fill the corresponding parameter correlation matrix unit, which is used to represent the structural constraint relationship between adjacent behavioral parameters during the high-pressure slurry valve seal recovery process.

[0014] Furthermore, the quantification of the behavioral offset includes: On the equivalent recovery process timeline, the complete recovery process is divided into several consecutive process sub-intervals; For each process sub-interval, the interval change of each recovery behavior parameter in the reference recovery behavior parameter set within that sub-interval is calculated, and the sign of the interval change is used as the reference change direction identifier for the corresponding recovery behavior parameter. For the second set of recovery behavior parameters in the current running state, the same process sub-interval division method is used to calculate the interval change of the corresponding recovery behavior parameter in each process sub-interval, and the sign of the interval change is used as the actual change direction identifier. When the actual change direction identifier of the same recovery behavior parameter is inconsistent with the reference change direction identifier within the same process sub-interval, it is determined that the recovery behavior parameter has deviated in direction within the process sub-interval. Within each process sub-interval, the number of recovery behavior parameters that deviate in direction is counted, and the ratio of the number to the total number of recovery behavior parameters included in the count within that sub-interval is defined as the direction deviation density of that process sub-interval. Based on the positional order of each process sub-interval on the equivalent recovery process time axis, a position weight function is introduced. By weighted summing of the deviation densities in each direction, the behavioral offset of the valve's current operating state relative to the reference state is obtained. The expression is: in, Indicates the number of associated terms that violate structural constraints; Indicates the first The direction of change for each associated item deviates from the identifier; These represent the ratios of the cumulative change magnitudes in the current state and in the reference state, respectively. These represent the equivalent process intervals in the current state and the reference state, respectively. The position weighting function is used to balance the contribution of different deviations. It changes monotonically as the equivalent recovery process progresses along the time axis, and is used to reduce the contribution of violation associations to the behavioral offset in the early stage of recovery and enhance the contribution of violation associations to the behavioral offset in the later stage of recovery.

[0015] Furthermore, the determination of the adaptive sealing state of the high-pressure slurry valve based on the change characteristics of behavioral offset includes: During multiple consecutive operation condition modulation processes, the corresponding behavior offset sequence is obtained. ,in Indicates the first The recovery process after the initial operational disturbance; Calculate the evolutionary increment between adjacent behavior offsets based on the behavior offset sequence. And construct the evolution trajectory of the behavioral offset; The evolution trajectory is segmented for consistency analysis to identify evolutionary segments in which the behavioral offset simultaneously satisfies a monotonically increasing relationship and the evolutionary increment does not decline within a preset number of consecutive perturbations. When the length of the evolution segment exceeds the preset minimum evolution length, and the behavioral offset within the evolution segment does not return to the stable interval corresponding to the sealing adaptive reference behavior model, the high-pressure slurry valve is determined to be in an internal leakage associated state.

[0016] Furthermore, the determination of the behavioral offset change characteristics is accomplished by performing irreversible evolutionary analysis on the behavioral offset sequence, the irreversible evolutionary analysis including: For continuously obtained behavioral offset sequences The system is divided into segments according to a preset sliding length, and the starting offset of each segment is determined. Section termination behavior offset and the minimum behavior offset within the segment ; Based on the distribution of behavioral offsets within the segment, calculate the irreversible evolution discriminant corresponding to the segment. The calculation formula is: in, This indicates the behavioral offset corresponding to the initial disturbance of the segment; This indicates the behavioral offset corresponding to the segment termination disturbance; This represents the minimum behavioral offset that occurs within the specified segment; When the irreversible evolution discriminant When the behavior offset exceeds the preset irreversible judgment threshold in multiple consecutive segments, and the behavior offset in the corresponding segment does not return to the allowable regression range defined by the sealed adaptive reference behavior model, the behavior offset change characteristics exhibit irreversible evolution characteristics.

[0017] Based on the above technical solution, the positive and beneficial effects of the present invention are as follows: 1. This invention does not rely on pressure, flow rate, or vibration characteristics at a single moment for judgment. Instead, it models the entire recovery behavior after the removal of operational disturbances, transforming the valve sealing structure's adjustment process to disturbances into a comparable recovery behavior path, and performing structured analysis on a unified equivalent recovery process time axis. This approach avoids misjudgments caused by operating condition fluctuations, changes in media characteristics, or short-term disturbances in existing methods. It ensures that the identification of internal leakage correlation states no longer relies on manual experience thresholds but is based on an objective judgment of whether structural shifts have occurred in the valve sealing adaptive behavior.

[0018] 2. This invention induces an observable recovery process at the valve sealing interface through operational condition modulation and makes a judgment based on the consistency of behavioral offset evolution in multiple disturbance responses. This allows it to capture the state characteristic of gradually decaying sealing adaptability before it manifests as a significant change in leakage. Compared to existing diagnostic methods that use internal leakage, acoustic signals, or temperature anomalies as triggering conditions, this invention can provide a judgment result when internal leakage is still in the correlation or evolution stage. This facilitates early detection of sealing failure trends and improves the foresight of high-pressure slurry valve operational status monitoring.

[0019] 3. This invention constructs a parameter correlation matrix by analyzing the direction, magnitude ratio, and evolution interval of changes in the recovery behavior parameters. It then analyzes the evolutionary characteristics of the behavior offset using an irreversible evolution discriminant, ensuring that the determination of the internal leakage correlation state is based on a stable structural offset that occurs repeatedly in multiple disturbance responses, rather than sporadic anomalies. This mechanism effectively suppresses false alarms caused by short-term load changes, fluctuations in medium particle distribution, or control disturbances, maintaining good consistency and repeatability of diagnostic results under long-term operation and complex working conditions. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort, wherein: Figure 1 This is a schematic diagram of the steps of the present invention; Figure 2This is a schematic diagram of the time-scale unification processing of the present invention; Figure 3 This is a schematic diagram illustrating the working principle of step four of the present invention; Figure 4 This is a schematic diagram illustrating the principle of determining the internal leakage correlation state in this invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0022] The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0023] Unless otherwise defined, all techniques and scientific methods used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The descriptions herein are for the purpose of illustrating particular embodiments only and are not intended to limit the invention. The terms "and / or" as used herein include any and all combinations of one or more of the associated listed items.

[0024] In one possible implementation, the application scenario is illustrated using a high-pressure slurry conveying pipeline in a mineral processing system. This high-pressure slurry conveying pipeline is used to transport high-solids-content slurry from the grinding section to the classification or flotation process. The pipeline's operating pressure is typically in the range of several megapascals. The transported medium contains a wide particle size distribution and high abrasiveness, placing high demands on the stability and adaptability of the valve sealing structure.

[0025] A high-pressure slurry valve is installed in the delivery pipeline. This valve adopts a structure in which the valve core and valve seat directly contact to form a sealing interface. The valve core is driven by an electro-hydraulic actuator. An upstream pressure sensor is installed in the upstream section of the valve, and a downstream pressure sensor is installed in the downstream section. Both pressure sensors are connected to the field control unit via analog signals. The valve actuator has a displacement feedback module inside, which is used to collect the axial displacement information of the valve core relative to the valve seat in real time. At the same time, the drive circuit of the actuator is equipped with a load feedback unit to reflect the changes in the driving force borne by the valve core under sealed contact conditions.

[0026] In terms of spatial arrangement, the upstream pressure sensor is installed on the valve inlet side at a distance of no more than three pipe diameters from the valve body, and the downstream pressure sensor is installed on the valve outlet side at a distance of no more than three pipe diameters from the valve body, in order to reduce the impact of pipeline pressure fluctuations on the diagnostic results. The displacement feedback module and load feedback unit are both integrated inside the valve actuator, without altering the original valve structure; this configuration does not constitute a limitation of the invention.

[0027] The aforementioned signals are transmitted to the industrial control unit via a fieldbus. This control unit can be an independent functional module within a PLC or DCS system, or it can be an additional embedded diagnostic controller. In this embodiment, the control unit internally includes an operating condition modulation module, a data acquisition and buffering module, a recovery behavior parameter calculation module, a behavior consistency analysis module, and a status determination module. These modules interact with each other via an internal data bus.

[0028] Under normal operating conditions, the high-pressure slurry valve is in a stable regulating state, and the sealing interface formed between the valve core and the valve seat is in a dynamic adaptive state under the action of high-pressure slurry and particles. Due to the continuous occurrence of microscopic processes such as particle embedding, migration, and rearrangement, the sealing interface is not static even in the absence of internal leakage.

[0029] Traditional diagnostic methods typically rely on abnormalities in flow rate or pressure across the valve, or changes in acoustic signals, to determine if internal leakage has occurred. However, under the aforementioned high-pressure slurry conditions, these methods are ineffective due to large system background fluctuations, strong signal noise, and the fact that early internal leakage has not yet formed a stable channel. The method described in this embodiment does not directly use leakage signals as the diagnostic object, but rather uses the recovery behavior of the valve sealing structure after controlled disturbances as the diagnostic basis.

[0030] During the system initialization phase, the control unit selects a time interval based on the operation records to confirm that the valve has not experienced internal leakage. This confirmation can be based on historical maintenance records, long-term operational stability assessments, or manual verification results. Within this time interval, the control unit activates the operating condition modulation module.

[0031] The operating condition modulation module sends modulation commands to the valve actuator to apply a limited-amplitude displacement modulation to the valve core without altering the system's process objectives. This modulation does not require a significant opening or closing action of the valve core; instead, it introduces a short-duration, controllable displacement change near the current valve position, with an amplitude less than the upper limit of the valve's normal adjustment accuracy range. The duration of this modulation process is shorter than the macroscopic adjustment cycle of the system's flow and pressure, thus avoiding interference with the process.

[0032] During and after modulation, the data acquisition and buffering module synchronously acquires upstream pressure signals, downstream pressure signals, valve core displacement signals, and actuator load signals at a fixed sampling period, and adds a uniform timestamp to each set of sampled data. The acquisition duration after modulation demodulation covers the complete process of the valve and pipeline system recovering from a disturbed state to a stable operating state.

[0033] The control unit marks the collected data as the first recovery behavior data and stores it in an internal buffer. Then, the recovery behavior parameter calculation module in the control unit processes the data in the buffer. Since the recovery process after the valve disturbance is removed has different durations in different operating cycles, directly comparing the original time series can easily introduce time scale bias. Therefore, before calculating the parameters, the first recovery behavior data is first processed to unify the time scale.

[0034] In this embodiment, the control unit uses the moment the disturbance is resolved as the start time of the recovery process and the moment the valve re-enters the steady-state operating range as the end time of the recovery process. Steady-state determination is achieved by jointly judging the rate of change of pressure before and after the valve and the rate of change of valve core displacement. When the aforementioned rates of change are all below a preset threshold for multiple consecutive sampling periods, the system is considered to have entered a steady state. The time window between the start and end times constitutes the original time window for a single recovery process.

[0035] To eliminate the impact of differences in recovery process durations on parameter comparisons, the control unit performs linear time mapping on the original time window, mapping it to a unified standard time interval. Specifically, let the original recovery time be... The standard recovery time is Then any time ttt on the original timeline is mapped to a time on the standard timeline. ,satisfy: The mapping process is completed at the software level through an interpolation algorithm to ensure that each operating parameter has a consistent number of sampling points and time positions within the standard time interval.

[0036] After unifying the time scale, the recovery behavior parameter calculation module extracts a set of recovery behavior parameters from the standardized time series to characterize the valve seal recovery path. This parameter set is not a simple statistical quantity, but a multi-dimensional parameter structure constructed specifically for the sealing characteristics of high-pressure slurry valves.

[0037] In this embodiment, the recovery behavior parameters include at least the following categories: First, the pressure recovery path parameter; by calculating the deviation of the pressure difference across the valve from the steady-state pressure difference, the pressure recovery integral is obtained, which characterizes the overall adjustment process experienced by the sealing interface in reforming an effective seal after disturbance. This parameter is obtained by integrating the pressure difference deviation over a standard time interval.

[0038] Second, valve core displacement recovery characteristic parameters; by statistically analyzing the distribution of the valve core displacement change rate during the recovery process, characteristic quantities reflecting the valve core's fine-tuning behavior during the sealing contact process are obtained. This parameter is used to characterize the dynamic adaptability of the sealing contact interface under particle participation conditions.

[0039] Third, the actuator load relaxation parameter; by quantifying the changes in the actuator drive load during the start and end phases of the recovery process, the load relaxation characteristics reflecting the changes in the contact state of the sealing interface are obtained. This parameter can reflect the intrinsic process of the contact state between the valve core and the valve seat transitioning from disturbance to stability.

[0040] The aforementioned multiple recovery behavior parameters together constitute the first set of recovery behavior parameters, and serve as the basic data for constructing the sealing adaptive reference behavior model.

[0041] During the reference model construction phase, the control unit summarizes and analyzes the set of first recovery behavior parameters obtained under multiple leak-free conditions. Since the recovery behavior under normal sealing conditions is not completely consistent but fluctuates within a certain range, the reference model does not use a single parameter value, but rather exists as a stable distribution region in the parameter space. This distribution region can be obtained statistically, for example, using the mean of each parameter and the allowable fluctuation range as the model boundary.

[0042] After the system is put into online operation, the control unit applies modulation disturbance to the high-pressure slurry valve in the online operating state according to the same operating condition modulation method as in the reference stage, and collects the operating parameters after the disturbance is removed to form the second recovery behavior data. This data is consistent with the first recovery behavior data in terms of processing flow, time scale unification method and parameter extraction method to avoid introducing additional system errors.

[0043] Subsequently, the behavior consistency analysis module compares and analyzes the second set of recovery behavior parameters with the sealing adaptive reference behavior model. In this embodiment, this comparison is not a simple threshold judgment, but is accomplished by constructing a behavior consistency metric function. This metric function comprehensively considers the degree of deviation of multiple recovery behavior parameters in the parameter space and reflects the differences in sensitivity of different parameters to the sealing state through a weighted approach.

[0044] The behavior offset output by the behavior consistency analysis module is a scalar value representing the overall deviation of the valve seal recovery behavior from the reference state under the current operating condition. This behavior offset is stored and appended with a time stamp to form a behavior offset evolution sequence.

[0045] During continuous operation, the state determination module periodically calls the behavior offset evolution sequence to analyze its changing trends. Unlike traditional diagnostic methods that are based on a single judgment, this embodiment uses a cross-cycle evolution criterion, that is, by comparing the direction of change and cumulative characteristics of behavior offset in multiple consecutive diagnostic cycles, it determines whether a systematic change has occurred in the sealing adaptive state.

[0046] When the behavior offset continuously shifts in the same direction over multiple cycles and gradually deviates from the stable distribution region of the reference model, the control unit determines that the high-pressure slurry valve has entered an internal leakage associated state. This state is not equivalent to an observable leak having occurred, but rather indicates that the adaptive adjustment capability of the valve sealing structure has shown a significant degradation trend.

[0047] As can be seen from the above implementation process, the present invention transforms the evolution process of the sealing interface, which was originally difficult to observe directly, into recoverable behavior parameters and behavior offsets that can be calculated, compared, and analyzed, thus changing the identification of internal leakage risk of high-pressure slurry valve from result-oriented to process-oriented.

[0048] In actual production operation, the diagnostic method of this invention does not involve the control unit continuously and frequently applying operating condition modulation to the high-pressure slurry valve. Instead, it executes the diagnostic process periodically through a scheduling strategy without affecting process stability. In this embodiment, the control unit sets the diagnostic cycle based on the operating load of the production system. For example, it triggers a diagnostic process when the system load changes little or when the process is in a steady-state range, thereby avoiding the superimposed effects of modulation operations and process adjustment commands.

[0049] Specifically, the control unit establishes a data interaction relationship with the field DCS system to obtain the current production cycle time, valve adjustment frequency, and pipeline pressure fluctuation level. When it is detected that the valve is in a relatively stable adjustment state and the system is not in a critical process switching phase, the operating condition modulation module is allowed to perform modulation operations. Before the modulation process is executed, the control unit performs a consistency check to ensure that the modulation amplitude, duration, and frequency are all within the preset safety range.

[0050] During long-term operation, the control unit assigns a unique diagnostic number to each diagnostic process and stores the corresponding set of second recovery behavior parameters, behavior offset, and judgment result in a structured manner. This data storage structure supports time-series queries, ensuring that the evolution trajectory of the behavior offset is completely preserved, thus forming a time-continuous record of the operational state evolution.

[0051] Considering that high-pressure slurry valves inevitably experience wear, changes in particle characteristics, and adjustments in operating conditions during long-term operation, this embodiment introduces a controlled update mechanism for the adaptive reference behavior model of the seal. Specifically, the control unit does not immediately correct the reference model when it detects a change in behavior offset, but only allows new recovery behavior parameters to be included in the reference model update range when preset stability conditions are met.

[0052] The stability condition includes at least the following constraints: Over multiple consecutive diagnostic cycles, the behavioral offset remains within the stable distribution region of the reference model and does not trigger an internal leakage-related state determination; simultaneously, the system operating parameters do not undergo structural changes, such as media type, operating pressure level, or valve structural parameters remaining unchanged. When the above conditions are met, the control unit gradually incorporates the new recovery behavioral parameters into the reference model through a weighted approach, enabling the model to adapt to slow evolution during long-term operation without masking abnormal trends.

[0053] At the software logic level, the behavior consistency analysis module and the state determination module are decoupled through clearly defined functional interfaces. The former focuses on quantitatively comparing recovery behavior parameters and outputting behavior offsets, while the latter is responsible for cross-cycle analysis and state determination of the behavior offset sequence. This design makes the diagnostic logic highly maintainable and scalable, and also facilitates adaptation and deployment on different models of high-pressure slurry valves.

[0054] Regarding hardware integration, the diagnostic method described in this embodiment does not require modification of the existing valve structure. The pressure, displacement, and load signals used are all operating parameters of the high-pressure slurry valve conventionally configured in industrial settings. The control unit can be deployed as an independent module in the valve's local control cabinet, or it can be integrated as a software function block into an existing DCS or PLC system, communicating with sensors and actuators via fieldbus or industrial Ethernet.

[0055] In actual operation, when the status determination module determines that the high-pressure slurry valve has entered an internal leakage associated state, it does not directly trigger an emergency shutdown or alarm action. Instead, it sends this state as operational risk information to the upper-level system. The upper-level system can then use this information to schedule maintenance plans, increase monitoring frequency, or adjust operating strategies, thereby managing potential internal leakage risks without affecting production continuity.

[0056] To facilitate a deeper understanding of the technology in this invention, a detailed description of a high-pressure slurry valve internal leakage diagnosis method disclosed in the embodiments of this application is provided below. Please refer to [link to relevant documentation]. Figure 1 The schematic diagram shown illustrates the steps of the invention, which include: Step 1: Apply an operational modulation disturbance to the high-pressure slurry valve in the state of no internal leakage confirmation using an operational condition modulation method to induce the redistribution of medium particles at the valve sealing interface. Collect system operating parameters before and after the operational disturbance to obtain first recovery behavior data representing the process of the valve entering a stable state after the disturbance is removed. Construct a sealing adaptive reference behavior model based on the first recovery behavior data. In actual industrial settings, high-pressure slurry valves typically operate for extended periods in high-pressure, high-solids-content conveying environments. The sealing interface between the valve core and seat not only withstands the pressure of the medium but also the embedding, rolling, and shearing effects of a large number of solid particles. There is a common understanding in the field that relative micro-movements between the valve core and seat should be minimized during operation to prevent accelerated wear of the sealing pair or the induction of early leakage. However, the inventors of this application, through long-term operational observation, have discovered that even in a leak-free state, the particle distribution within the sealing interface is not static but rather exists in a slowly evolving "adaptive equilibrium" state. This state, once slightly disturbed, exhibits a recovery process that itself contains crucial information about the sealing performance.

[0057] Based on the above understanding, this step introduces operating condition modulation as an active excitation method. In one possible implementation, operating condition modulation is achieved by applying periodic micro-displacement control to the actuator of the high-pressure slurry valve. This micro-displacement control is not a traditional valve opening adjustment, but a small reciprocating displacement around the current working valve position. The displacement amplitude is limited to a safe range that does not cause changes in the valve's on / off state or a step change in the process flow rate.

[0058] It should be noted that the "micro-displacement control" in this application is different from valve opening and closing control. Its purpose is not to change process parameters, but to disturb the local contact state between the valve core and the valve seat. At the same time, "operating condition modulation" is also different from process load fluctuation. Its modulation object is the sealing interface structure rather than the conveying capacity.

[0059] In practical implementation, the micro-displacement amplitude can be preset according to the valve specifications, sealing structure, and actuator resolution. For example, in a high-pressure slurry valve using a linear stroke valve core structure, the micro-displacement amplitude can be controlled within one-thousandth of the rated stroke. After each micro-displacement, the system maintains this displacement state for a preset holding time. This holding time is not arbitrarily set; its technical purpose is to provide sufficient time for the slurry particles at the sealing interface to rearrange their positions, release stress, or re-embed.

[0060] As one possible implementation, if the holding time is too short, the particles will enter the next displacement before completing the rearrangement, resulting in the disturbance effect being "averaged" and making it impossible to form a distinguishable recovery process. If the holding time is too long, the disturbance effect tends to dissipate, which is not conducive to the identification of subsequent recovery behavior. Therefore, the holding time can be set empirically or adjusted online based on the particle size distribution of the medium, the viscosity of the slurry, and the operating pressure.

[0061] After completing a predetermined number of operating condition modulations and then demodulating, the valve enters a natural recovery phase. This step involves continuously collecting system operating parameters during this phase to generate first recovery behavior data. Unlike traditional methods that only focus on steady-state parameters, this application specifically focuses on the dynamic recovery process after disturbance removal. In a typical implementation, the first recovery behavior data includes at least the upstream pressure change sequence, the downstream pressure change sequence, and the valve position feedback change sequence.

[0062] It should be noted that the reason for choosing the above parameter combination in this application is that the pressure before and after the valve can indirectly reflect the change in the influence of the sealing interface on the flow resistance, while the valve position feedback is used to eliminate the interference of the actuator control error on the pressure change, thereby ensuring that the observed recovery behavior mainly comes from the adaptive adjustment of the internal structure of the sealing interface.

[0063] In terms of time processing, all recovery behavior data are aligned with the moment when the operating condition modulation is released as a unified time reference point. Subsequently, based on the parameter change rate, fluctuation amplitude, and evolution trend, the recovery process is divided into at least one unsteady-state recovery segment and one steady-state approaching segment. The unsteady-state recovery segment typically corresponds to the rapid rearrangement stage of the internal particle structure of the sealing interface, while the steady-state approaching segment corresponds to the stage where a new contact equilibrium gradually forms.

[0064] Based on the above segmentation results, a sealed adaptive reference behavior model is constructed. This model is not a single function or a fixed threshold, but a behavioral description structure jointly defined by multiple constraints. Among them, the stage sequence constraint is used to limit the order in which each recovery segment appears on the time axis; the stage connection constraint is used to limit the continuous evolution characteristics between adjacent segments to prevent abrupt changes that do not conform to the physical process; and the stability constraint is used to limit the allowable natural fluctuation range of system parameters within the steady-state approaching segment.

[0065] It should be noted that the "Sealing Adaptive Reference Behavior Model" in this application differs from the traditional static benchmark model. Essentially, it is a dynamic behavioral reference based on the recovery path, used to describe the normal response of the sealing interface to disturbances under conditions of no internal leakage. This model provides a reliable benchmark for subsequent comparative analysis of recovery behavior under online operating conditions, thus making internal leakage diagnosis no longer dependent on the occurrence of overt leakage phenomena.

[0066] Step 2: Apply the same or equivalent operating condition modulation method as in Step 1 to the high-pressure slurry valve in online operation, and collect the system operating parameters after the operating disturbance is removed to obtain the second recovery behavior data of the valve in the current operating state. The core of this step is to induce and capture the instantaneous recovery behavior of the valve sealing interface to controllable disturbances without interfering with normal process delivery or introducing additional operating risks, thereby providing a comparable data basis for subsequent behavior consistency analysis.

[0067] Unlike step one, step two deals with a valve that is not in a specially confirmed leak-free baseline state, but rather in a continuous material flow operation state during actual production. In this state, the pressure load, flow shocks, and minor control fluctuations of the actuator can all affect the disturbance response process. Therefore, before triggering the operational condition modulation, this step first assesses the current operational stability of the valve.

[0068] In one possible implementation, the system continuously monitors the valve position feedback signal of the high-pressure slurry valve and calculates the change amplitude of the valve position feedback based on a preset time window. When the change amplitude is within a preset allowable range, the system determines that the current operating state of the valve is within an acceptable stable range, thereby allowing the triggering of operating condition modulation that is the same as or equivalent to step one. The technical significance of this discrimination mechanism is to avoid superimposing new modulation actions when the valve actuator is frequently adjusted or affected by external disturbances, thereby reducing the impact on operational safety.

[0069] It should be noted that the "operational stability judgment" in this application is different from the overall stability assessment of the process system. Its focus is on the execution state of the valve body, used to determine whether it is appropriate to apply disturbance to the sealing interface, rather than to determine whether there is a process abnormality in the conveying system.

[0070] After determining that the operational stability conditions are met, the system triggers the operational condition modulation method corresponding to step one, applying periodic micro-displacement disturbances to the valve. This modulation method maintains consistency with step one in terms of displacement amplitude, action rhythm, and holding method, thereby ensuring that the subsequent recovery behavior is physically comparable.

[0071] After the operating condition modulation is released, the system synchronously acquires the upstream pressure, downstream pressure, and valve position feedback signals of the valve. The purpose of synchronous acquisition is to preserve the temporal correlation between the parameters and avoid misjudgment of the recovery path characteristics due to sampling timing deviations. In actual implementation, multiple channels of signals can be timestamped using the same data acquisition unit or a unified clock source.

[0072] Because operational disturbances are unavoidable in the online operating environment, this step does not directly use all collected data as the data for the second recovery action. Instead, it performs segment filtering on the collected data. Specifically, firstly, data segments before the modulation of the operating conditions is removed are eliminated to avoid introducing the modulation process itself into the recovery analysis; secondly, data segments where the parameters show non-monotonic changes after the disturbance is removed are eliminated.

[0073] It should be noted that "non-monotonic change" in this application does not mean that the parameters must change strictly monotonically, but rather that there is a clear reverse trend, multiple rebounds or abnormal oscillations during the recovery process. Such changes usually reflect external operating condition disturbances or the compensation behavior of the control system, rather than the recovery characteristics of the sealing interface itself.

[0074] After completing the above screening, only data segments that simultaneously meet the monotonic recovery characteristics and whose duration exceeds the preset minimum observation time are constructed as the second recovery behavior data. The purpose of setting the minimum observation time is to ensure that the selected data segments can fully cover the main stages of the evolution of the sealed interface from the perturbed state to the stable state, and to avoid missing key recovery characteristics due to too short an observation time.

[0075] The second recovery behavior data obtained through the above method is consistent with the first recovery behavior data in step one in terms of data structure, time base and parameter dimensions, thus providing reliable input for the comparative analysis based on behavior consistency in subsequent steps.

[0076] Step 3: Perform time-scale unified processing on the first recovery behavior data and the second recovery behavior data, and extract a set of recovery behavior parameters based on the processed data to reflect the recovery path characteristics of the valve after disturbance. The set of recovery behavior parameters is used to represent the adjustment behavior of the valve sealing structure to the current operating disturbance. In the actual operating environment of high-pressure slurry valves, the recovery process after the removal of operational disturbances is affected by various factors such as slurry concentration, flow rate, upstream pressure, and actuator status. Even if the disturbance is exactly the same, the recovery process under different times and operating conditions may differ significantly on the physical time scale. For example, under high load conditions, valve parameters may take a long time to stabilize, while under low load conditions, the recovery process may be extremely rapid. If the two recovery processes are directly compared based on physical time, the conclusions often reflect the differences in operating conditions rather than the differences in the state of the sealing structure itself.

[0077] Therefore, this embodiment does not directly compare the first recovery behavior data and the second recovery behavior data on the physical timeline. Instead, it first performs time-scale unification processing on the two types of data. This processing is not a simple time stretching or compression, but rather revolves around the physically defined process variable of the recovery process.

[0078] Please see Figure 2 The schematic diagram shown illustrates a possible implementation method whereby, firstly, the complete recovery process of the system operating parameters from the moment the operational disturbance is resolved to the point where the system enters the stable state determination interval is determined, based on the first recovery behavior data and the second recovery behavior data, respectively. The moment the operational disturbance is resolved is denoted as... This serves as the unified starting point for the recovery process; the moment when the steady-state determination interval is entered is denoted as... This is used to define the termination point of the recovery process.

[0079] It should be noted that the "stable state determination interval" in this application is different from the instantaneous parameter constant state. It refers to the state in which the change of system operating parameters within a certain time window is continuously within a preset threshold range, and is used to characterize the valve entering a new mechanical equilibrium interval after disturbance.

[0080] After determining the recovery process interval, at least one parameter related to the change in load on the valve sealing interface is selected from the system operating parameters as a characterization of the recovery process. In practice, this parameter can be selected from one or more of the valve's upstream pressure, downstream pressure, or valve position feedback. The reason for choosing this type of parameter is that its changes directly or indirectly reflect the contact load and fluid interaction state experienced by the sealing interface.

[0081] Subsequently, based on the rate of change of the selected recovery process characterization quantities, the physical time axis is nonlinearly reconstructed to obtain the equivalent recovery process time axis. This equivalent recovery process time axis is defined by the following relationship: in, Indicates the time when the operational disturbance is resolved; This indicates the moment when the system operating parameters enter the stable state determination interval; Represents any physical time sampling point during the recovery process; For integration variables; Indicates at time The acquired recovery process characteristics; This represents the instantaneous rate of change amplitude of the characterization quantity during the recovery process; The normalized equivalent recovery process time variable has a value range of [0,1].

[0082] The above definition normalizes the cumulative degree of "variable activity" during the recovery process, mapping the originally non-alignable recovery process in physical time to a unified process scale. At this scale, τ=0 corresponds to the state where the disturbance has just been removed, τ=1 corresponds to the state where the system has completed the main recovery and entered the stable region, and any value in between reflects the relative progress of the recovery process at the level of sealing regulation behavior.

[0083] It should be noted that the absolute value integral of the rate of change is introduced in this application, rather than directly integrating with respect to the parameter itself, in order to avoid the cancellation effect of different positive and negative change directions of the parameter on the process characterization, thereby more realistically reflecting the "activity intensity" in the recovery process.

[0084] After obtaining the equivalent recovery process timeline, the first recovery behavior data and the second recovery behavior data are mapped onto the equivalent timeline and then uniformly resampled. Through this process, the recovery processes, which originally had different lengths and rhythms in the physical time dimension, are transformed into data sequences with consistent resolution in the same process dimension.

[0085] After completing the unified resampling, a set of recovery behavior parameters reflecting the overall shape of the recovery process is extracted based on the recovery path obtained from the resampling. In practice, this parameter set includes at least the following three types of parameters: recovery process slope distribution parameters, used to characterize the advancement rate characteristics of different stages in the recovery process; recovery path curvature parameters, used to describe whether there are obvious inflection points or nonlinear adjustment behaviors in the recovery process; and steady-state region precursor evolution length parameters, used to characterize the length of the effective adjustment process experienced by the system before entering a steady state.

[0086] The parameters mentioned above are not simple statistics, but are used to describe the regulating behavior of the valve sealing structure to operational disturbances from the perspective of process morphology, providing a clear physical meaning for subsequent comparative analysis based on behavioral consistency.

[0087] Step 4: Use the behavior consistency analysis method to compare the set of recovery behavior parameters with the sealing adaptive reference behavior model to obtain the behavior offset of the valve's current operating state relative to the reference state; In practice, step four runs as an independent behavior consistency analysis module in the control system or edge computing unit, used to compare and analyze the recovery behavior structure of the valve in its current operating state with that of the reference state. The input to this module is the set of recovery behavior parameters output from step three and their identification results on the equivalent recovery process time axis, and the output is the behavior offset characterizing the degree of deviation of the valve's operating state.

[0088] Please see Figure 3 The schematic diagram illustrates that in actual engineering deployments, the set of recovery behavior parameters is first organized into structured data objects. Each recovery behavior parameter includes the following information: parameter type identifier, parameter value, the start and end positions of the corresponding equivalent recovery process time interval, and the dominant stage number of the parameter in the recovery process. This structured representation avoids analysis based solely on a single numerical value, thus preserving the phased information of the recovery process.

[0089] To construct an operational expression for the sealing adaptive reference behavior model, after step one, the system statistically organizes the set of recovery behavior parameters obtained under multiple leak-free baseline conditions. Specifically, for each type of recovery behavior parameter, its occurrence stage on the equivalent recovery process time axis is discretized into several standard process segments, and a statistical description is established for the parameter behavior within each process segment, including the set of allowed change directions, the upper and lower bounds of the change amplitude, and the typical evolution interval range between adjacent process segments. The above statistical results are stored in matrix form, forming a reference parameter correlation matrix.

[0090] It should be noted that the "parameter correlation matrix" in this application is not a simple numerical matrix, but a composite constraint matrix, where each matrix element corresponds to a constraint relationship description between a pair of adjacent recovery behavior parameters. This constraint relationship includes at least three types of information: directional consistency constraints, amplitude range constraints, and process interval constraints. In this way, the inherent structural relationships between recovery behaviors are transformed into a clear data structure, facilitating subsequent programmatic processing.

[0091] In practical implementation, the parameter correlation matrix is ​​first based on the set of recovery behavior parameters after time-scale unification processing in step three, and each recovery behavior parameter is sorted and numbered on the equivalent recovery process time axis. It should be noted that the "recovery behavior parameter" in this application is not the parameter value of an instantaneous sampling point, but an abstract expression of the behavioral characteristics within a specific process interval during the recovery process, which usually corresponds to a sub-interval of the recovery path with a clear monotonic change trend.

[0092] For any two recovery behavior parameters that are adjacent on the equivalent recovery process time axis and The system is in its corresponding recovery process interval The three types of correlation quantities used to characterize the structural relationship between the two are calculated separately.

[0093] First, the monotonic change direction identifier is calculated. This identifier is obtained by analyzing the first-order sign of the recovery behavior parameter within its corresponding process interval. In engineering implementation, the first-order difference or numerical differentiation of the parameter's change curve with the equivalent recovery process time is typically performed, and sign consistency is judged throughout the entire process interval. When the parameter shows an overall upward trend, the direction identifier is recorded as positive; when it shows an overall downward trend, it is recorded as negative; if the direction reverses frequently within the interval, the parameter pair is considered to not satisfy the monotonicity requirement within that interval.

[0094] It should be noted that the "direction of monotonic change" in this application is different from the "sign of instantaneous slope" in traditional signal processing. It focuses on the overall evolution direction within the recovery process interval and is used to reflect the dominant regulatory trend of the sealing interface in this stage.

[0095] Secondly, calculate the ratio of cumulative change amplitude. This ratio is not a direct representation of the ratio of parameter changes, but rather obtained by integrating the magnitude of the parameter change rate over the process interval. Specifically, it is obtained by integrating the magnitude of the parameter change rate over the process interval. and In respectively Integrating over the interval yields the cumulative change intensity of the two parameters during the recovery phase, and the ratio of the two is then used as the cumulative change amplitude ratio. .

[0096] Based on the above definition, even if the net changes of two parameters are similar, the intensity and rhythm of their adjustment during the recovery process may still differ significantly. The ratio of cumulative change amplitudes can effectively characterize the adjustment emphasis of the sealing structure on different physical quantities at this stage. It should be noted that the domain of this ratio is positive real numbers, and its physical meaning does not depend on the specific dimensions of the parameters, thus providing a basis for structural comparisons between different types of parameters.

[0097] Next, calculate the equivalent process interval. This quantity is directly determined by the difference in position between adjacent recovery behavior parameters on the equivalent recovery process time axis, i.e. Since the equivalent recovery process timeline has eliminated the differences in physical time scales, this interval reflects the stage span of the recovery behavior at the process level, rather than an absolute time interval.

[0098] In engineering practice, this process interval is used to characterize the "recovery phase length" experienced by the sealing interface as it transitions from one adjustment behavior to the next. It is of great significance for identifying whether the sealing structure has abnormal hysteresis or premature failure.

[0099] The above-mentioned monotonic change direction identifier and cumulative change amplitude ratio and equivalent process interval They are combined into a single associated element and filled into the corresponding parameter association matrix. and Within the matrix cells, the constructed parameter correlation matrix fully records the structural constraint relationships between adjacent recovery behavior parameters during the high-pressure slurry valve seal recovery process.

[0100] After the valve is in online operation and steps two and three are completed, the system will number and structure the set of recovery behavior parameters obtained under the current operating state according to the same rules as the reference model, generating an actual recovery behavior sequence. Each parameter node in this sequence can find its corresponding theoretical constraint description in the reference parameter association matrix.

[0101] Subsequently, the behavior consistency analysis module compares the actual parameter correlation matrix with the reference parameter correlation matrix item by item. During the comparison, it doesn't simply determine whether a parameter exceeds its limits, but rather examines the evolutionary relationship of each pair of adjacent recovery behavior parameters. For example, for a pair of parameters that are required to maintain a consistent direction of change in the reference model, the system determines whether the direction of change of this parameter pair has reversed or separated under actual operating conditions; for amplitude range constraints, it determines whether the actual change continuously exceeds the reference range; for process interval constraints, it compares whether the distance between corresponding stages of adjacent parameters on the equivalent recovery process time axis has abnormally compressed or stretched.

[0102] It should be noted that this invention specifically introduces a "persistence determination" mechanism. That is, a constraint violation is only recorded as a valid violation associated item when it crosses a preset minimum process proportion in the equivalent recovery process, thereby avoiding interference from instantaneous noise or local disturbances on the calculation of behavioral offset.

[0103] After comparing all related items, the system statistically distributes the constraint-violation related items according to their position on the equivalent recovery process timeline, and performs a weighted calculation based on the deviation magnitude of each related item to generate a behavioral offset. This behavioral offset essentially reflects the overall deviation of the recovery behavior structure from the reference model, rather than the anomaly of a single parameter.

[0104] In the specific implementation process, for each associated item that violates the structural constraints, the system quantifies its deviation degree from three dimensions: direction of change, distribution of adjustment intensity, and distribution of recovery process. These correspond to "directional consistency," "energy or load distribution relationship," and "stage rhythm characteristics" in the adjustment behavior of the sealed structure, respectively, and together they constitute the basic structural unit of the sealed adaptive behavior.

[0105] Among them, the direction of change deviates from the label. This is used to reflect whether the overall evolution direction of adjacent recovery behavior parameters is consistent with the reference state throughout the entire recovery process interval. It should be noted that the "direction of change" in this application does not refer to the instantaneous slope at a specific moment, but is determined based on the statistical results of the first-order change sign within the equivalent recovery process interval, to avoid interference from local perturbations or measurement noise in the direction determination. When the directional relationship in the actual state is inconsistent with the reference state, It is assigned a non-zero label to characterize the mismatch of regulatory behavior at the trend level.

[0106] The amplitude ratio deviation is used to characterize whether the relative relationship of the cumulative change intensity of adjacent recovery behavior parameters changes during the recovery process. In engineering implementation, it is calculated by taking the cumulative change amplitude ratio under actual conditions. and corresponding to the reference state By comparison, we can obtain This deviation does not focus on whether the change of a single parameter increases or decreases, but rather on "whose regulatory contribution to whom has changed," which is particularly sensitive when there is particle embedding, uneven wear, or redistribution of contact stress at the sealing interface of a high-pressure slurry valve.

[0107] Process interval deviation This describes whether the phase distribution of the recovery behavior on the equivalent recovery process time axis is stretched or compressed. It should be noted that the "equivalent recovery process interval" in this application differs from the physical time interval; it reflects the relative positional change of the recovery behavior at the adjustment process level. When the adaptive capability of the sealing interface decreases, it often manifests as the premature termination or delayed occurrence of certain recovery phases. Such changes may be masked by differences in operating conditions on the physical time axis, but can be stably identified on the equivalent recovery process axis.

[0108] After obtaining the above three types of deviations, a position weight function is introduced. This is used to reflect the difference in importance of the location of different behavioral offsets to the overall sealing status. As one possible implementation, based on engineering experience or historical statistics, offsets closer to the initial recovery stage can be given higher weight to highlight changes in the initial adjustment capability of the sealing interface; alternatively, when steady-state sealing performance needs to be considered, the weight ratio of offsets closer to the steady-state approach can be increased. The specific form of the weighting function can be set according to the valve type, slurry characteristics, and operating strategy, and is not limited thereto.

[0109] Finally, the behavioral offset D is obtained by weighting and summing all associated terms that violate structural constraints, and its expression is: Where N represents the number of associated terms that violate structural constraints; α, β, and γ are weighting coefficients used to balance the contribution of different types of deviations, and their values ​​can be set according to the sensitivity requirements for mismatch in change direction, mismatch in adjustment intensity, or mismatch in process rhythm.

[0110] It is important to emphasize that the aforementioned behavioral offset is not something that can be obtained through conventional logical reasoning as in existing technologies. Existing diagnostic methods are typically based on parameter thresholds, statistical distributions, or overall similarity evaluations, which cannot identify the disruption of the internal structural relationships of the recovery behavior. This application introduces a screening mechanism of "correlation violation" and a quantification method of "process position weighting," enabling the behavioral offset to directly reflect the degradation characteristics at the sealing adaptive mechanism level, thereby providing a stable, interpretable, and engineering-significant quantitative basis for subsequent determination of internal leakage correlation status.

[0111] Step 5: Determine the sealing adaptive state of the high-pressure slurry valve based on the changing characteristics of the behavioral offset. When the behavioral offset meets the preset evolution conditions, the high-pressure slurry valve is determined to be in an internal leakage associated state. It should be noted that the "internal leakage associated state" in this application is different from the "internal leakage state" commonly referred to in the prior art. The internal leakage associated state does not require that the valve has already experienced observable media penetration or external leakage, but rather refers to the fact that the adjustment capability of the sealing structure has deviated from its reference adaptive behavior model and no longer has the ability to return to a healthy state in subsequent operation. In engineering practice, this state often appears before obvious internal leakage.

[0112] Please see Figure 4 The framework diagram shown illustrates that, in actual implementation, the system does not immediately determine internal leakage after a single operational condition modulation. Instead, it periodically or on demand triggers multiple operational condition modulations during the normal continuous operation of the high-pressure slurry valve. After each modulation, the corresponding behavioral offset is obtained according to the aforementioned steps. This results in a sequence of behavioral offsets ordered by the number of perturbations. The number of perturbations here It is not a sampling sequence number in the sense of physical time, but refers to the number of comparable disturbances applied to the same valve sealing structure under the same or equivalent operating conditions.

[0113] It should be noted that the "behavioral offset sequence" in this application is different from traditional time series data. Its sequence dimension reflects the "state transition path of the regulatory capacity after the disturbance", rather than the natural fluctuation of the operating parameters over time.

[0114] After obtaining the behavior offset sequence, the system first calculates the behavior offset evolution increment corresponding to two adjacent perturbations. This evolutionary increment reflects the changing trend of valve sealing regulation behavior under the same disturbance intensity and modulation method. If A positive value indicates that the sealing structure's deviation from disturbances is increasing; if... A negative value indicates a certain degree of self-recovery trend.

[0115] Based on this, the system does not directly rely on Instead of judging the sign, it constructs an evolutionary trajectory for the behavioral offset sequence and performs segmented consistency analysis on this trajectory. As a possible implementation, a fixed-length or adaptive-length perturbation window can be used to analyze multiple consecutive perturbation cycles. The consistency of the sign and the change in amplitude are comprehensively judged to identify whether there is an evolutionary segment that remains monotonically increasing within a preset number of consecutive disturbances and does not show a significant decline.

[0116] It is important to emphasize that "monotonically increasing" does not mean point-by-point increase in a strict mathematical sense, but rather that within an allowable range of small fluctuations, the overall evolution trend does not reverse. This setting is used to accommodate unavoidable measurement noise and transient disturbances in actual operating conditions.

[0117] When the length of the identified evolutionary segment exceeds the preset minimum evolutionary length, and all behavioral offsets within that segment fail to return to the stable range defined by the sealing adaptive reference behavior model, the system initially determines that the valve sealing structure has deviated from the healthy adaptive state.

[0118] To further avoid misjudgments caused by short-term cumulative effects or occasional changes in operating conditions, this application introduces an irreversible evolution discriminant. The "irreversibility" of the behavioral offset evolution process is quantitatively characterized. In specific implementation, the behavioral offset sequence is divided into sliding segments, and each sliding segment corresponds to a set of behavioral offset quantum sequences after continuous perturbation.

[0119] Within each sliding segment, the system determines the initial offset of the segment. Section termination behavior offset and the minimum behavioral offset occurring within the segment It should be noted that, It is not limited to the beginning or end of the segment; it is used to characterize whether there is a clear recovery attempt throughout the entire segment.

[0120] Based on the above quantities, an irreversible evolution discriminant is introduced: The physical meaning of this discriminant is that when the behavioral offset within a segment fluctuates locally, but the overall evolution always progresses in a direction away from the reference state, the denominator and numerator are close. Approaching 1 or greater; while when there is a significant decline within the segment, i.e., the behavioral offset was significantly close to the reference state, then near Thus Significantly reduced.

[0121] It should be noted that "irreversible evolution" in this application is different from simple "continuous increase". Irreversible evolution emphasizes that after experiencing multiple disturbances, the system has lost its ability to return to the stable range of the reference behavior model. In engineering, this degradation of ability usually means that the particle embedding state of the sealing interface, the contact stress distribution, or the micro-contact morphology has undergone structural changes.

[0122] In the actual judgment process, when the irreversible evolution discriminant... When the deviation exceeds the preset irreversible threshold in multiple consecutive sliding sections, and the behavior offset in the corresponding section does not enter the regression allowable range defined by the sealing adaptive reference behavior model, the system finally determines that the high-pressure slurry valve is in an internal leakage associated state.

[0123] It should be noted that the above-mentioned judgment logic cannot be obtained through a single threshold comparison or statistical trend analysis as in existing technologies. This application, by structurally modeling the evolution path, fallback capability, and stage consistency of behavioral offsets, transforms the internal leakage judgment from "outcome identification" to "capability degradation identification," significantly improving the reliability of judgment in the early stages of internal leakage.

[0124] Finally, it should be noted that the mathematical formulas, derivations, symbol definitions, and parameter calculation methods used in this specification are all for the purpose of further clarifying and verifying the technical content of this invention, so that those skilled in the art can more intuitively and accurately understand the working mechanism and technical effects of this invention. These formulas are only used as quantitative expressions or illustrative examples of technical features and do not constitute limiting conditions of the claims of this invention. Those skilled in the art should understand that, without changing the core idea of ​​this invention, the parameter forms, calculation methods, numerical ranges, and even symbol representations involved in the formulas can be equivalently replaced or simplified in engineering according to the actual application environment. The specifics can be determined according to the actual situation, and no limitation is imposed. It should also be emphasized that the formulas in this specification are not theoretical derivations in the style of academic research papers, but rather an engineering description of the embodiments of this invention. Their purpose is to enhance the understandability and implementability of this invention, rather than to increase redundancy and complexity. Those skilled in the art can choose whether to use such quantitative tools when reading this specification, or can achieve the same technical effects through other equivalent methods.

[0125] Furthermore, while specific embodiments of the present invention have been described above, those skilled in the art should understand that these specific embodiments are merely illustrative. Those skilled in the art can omit, substitute, and modify the details of the above methods and systems in various ways without departing from the principles and essence of the present invention. For example, combining the above method steps to perform substantially the same function and achieve substantially the same result according to substantially the same method falls within the scope of the present invention. Therefore, the scope of the present invention is defined only by the appended claims.

Claims

1. A method for diagnosing internal leakage in a high-pressure slurry valve; characterized in that: Includes the following steps: Step 1: Apply an operational modulation disturbance to the high-pressure slurry valve in the state of no internal leakage confirmation using an operational condition modulation method to induce the redistribution of medium particles at the valve sealing interface. Collect system operating parameters before and after the operational disturbance to obtain first recovery behavior data representing the process of the valve entering a stable state after the disturbance is removed. Construct a sealing adaptive reference behavior model based on the first recovery behavior data. Step 2: Apply the same or equivalent operating condition modulation method as in Step 1 to the high-pressure slurry valve in online operation, and collect the system operating parameters after the operating disturbance is removed to obtain the second recovery behavior data of the valve in the current operating state. Step 3: Perform time-scale unified processing on the first recovery behavior data and the second recovery behavior data, and extract a set of recovery behavior parameters based on the processed data to reflect the recovery path characteristics of the valve after disturbance. The set of recovery behavior parameters is used to represent the adjustment behavior of the valve sealing structure to the current operating disturbance. Step 4: Use the behavior consistency analysis method to compare the set of recovery behavior parameters with the sealing adaptive reference behavior model to obtain the behavior offset of the valve's current operating state relative to the reference state; Step 5: Determine the sealing adaptive state of the high-pressure slurry valve based on the change characteristics of the behavior offset. When the behavior offset meets the preset evolution conditions, determine that the high-pressure slurry valve is in an internal leakage associated state.

2. The method for diagnosing internal leakage in a high-pressure slurry valve according to claim 1, characterized in that: The operating condition modulation method includes applying periodic micro-displacement control to the valve core of the high-pressure slurry valve relative to the valve seat. The micro-displacement control range is within the displacement interval that does not change the valve's on / off state, and a preset holding time is maintained after each micro-displacement control, inducing the slurry particles at the valve seat sealing surface to rearrange their positions in the contact area.

3. The method for diagnosing internal leakage in a high-pressure slurry valve according to claim 1, characterized in that: The first recovery behavior data consists of a sequence of pressure changes before the valve, a sequence of pressure changes after the valve is removed, and a sequence of valve position feedback changes, all continuously collected after the modulation disturbance is removed. The first recovery behavior data uses the moment the disturbance is removed as a unified time reference point and is divided into at least one unsteady recovery segment and one steady-state approaching segment according to the evolution characteristics of the system operating parameters. This is used to construct a sealing adaptive reference behavior model to describe the staged recovery behavior of the valve sealing interface during particle redistribution. The sealing adaptive reference behavior model includes stage sequence constraints to limit the order of occurrence of each recovery segment, stage connection constraints to limit the evolution continuity of adjacent recovery segments, and stability constraints to limit the range of system parameter fluctuations within the steady-state approaching segment.

4. The method for diagnosing internal leakage in a high-pressure slurry valve according to claim 1, characterized in that: The acquisition of the second recovery behavior data includes: When the high-pressure slurry valve is in continuous feeding operation, the valve's current operating stability is judged based on the valve position feedback signal. When the valve position feedback change is within the preset allowable range, the operating condition modulation corresponding to step one is triggered. After the modulation of the operating condition is released, the upstream pressure, downstream pressure and valve position feedback signal of the valve are collected synchronously. The collected data is segmented and the data segments before the modulation of the operating condition is removed and the data segments whose parameters show non-monotonic changes after the disturbance is removed are removed. Data segments that satisfy the monotonic recovery characteristics and whose duration exceeds the preset minimum observation time are constructed as the second recovery behavior data of the valve under the current operating state.

5. The method for diagnosing internal leakage in a high-pressure slurry valve according to claim 1, characterized in that: The time-scale unification processing of the first recovery behavior data and the second recovery behavior data includes: The recovery process of the valve from the moment the operational disturbance is removed to the time interval for determining the stable state is determined in the first recovery behavior data and the second recovery behavior data, respectively; Select at least one of the valve upstream pressure data, valve downstream pressure data, and valve position feedback data as a characterization of the recovery process; Based on the rate of change of the recovery process characteristic quantities during the recovery process, the physical time axis is nonlinearly reconstructed to obtain the equivalent recovery process time axis. The reconstruction expression is: in, Indicates the time when the operational disturbance is resolved; This indicates the moment when the system operating parameters enter the stable state determination interval; Represents any physical time sampling point during the recovery process; Represents the integral variable; Indicates at time The collected system operating parameters are selected from at least one of valve upstream pressure, downstream pressure, or valve position feedback. This represents the magnitude of the instantaneous rate of change of the system's operating parameters during the recovery process; This represents the normalized equivalent recovery process time variable, with a value range of [value missing]. This is used to characterize the evolution of valve disturbance recovery at the level of sealing regulation behavior; On the equivalent recovery process timeline, the first recovery behavior data and the second recovery behavior data are uniformly resampled; Based on the resampled recovery path, a set of recovery behavior parameters is extracted from the second recovery behavior data. The set of recovery behavior parameters includes at least the recovery process slope distribution parameter, the recovery path curvature parameter, and the stable region precursor evolution length parameter.

6. The method for diagnosing internal leakage in a high-pressure slurry valve according to claim 1, characterized in that: The process of comparing the second set of recovery behavior parameters with the sealing adaptive reference behavior model using the behavior consistency analysis method includes: Based on the sealing adaptive reference behavior model, each recovery behavior parameter in the recovery behavior parameter set is numbered according to its order of appearance on the equivalent recovery process time axis to construct a reference recovery behavior sequence; The direction of change, range of change, and evolution interval of adjacent recovery behavior parameters in the reference recovery behavior sequence are statistically analyzed to generate a parameter correlation matrix that characterizes the intrinsic constraint relationship of recovery behavior under the reference state. Based on the set of recovery behavior parameters, the actual recovery behavior sequence under the current running state is constructed according to the same numbering rule, and the corresponding actual parameter association matrix is ​​generated; The actual parameter correlation matrix is ​​compared with the reference parameter correlation matrix item by item to identify correlation items that violate the constraints of change direction, amplitude range, or evolution interval. Based on the distribution density and deviation magnitude of the violation associated terms on the equivalent recovery process time axis, the behavioral offset of the valve's current operating state relative to the reference state is quantitatively characterized.

7. The method for diagnosing internal leakage in a high-pressure slurry valve according to claim 6, characterized in that: The construction of the parameter correlation matrix includes: On the equivalent recovery process timeline, for any two adjacent recovery behavior parameters and The monotonic change direction identifier, cumulative change amplitude ratio, and equivalent process interval are calculated respectively within the recovery process interval; wherein, the monotonic change direction identifier is determined by the first-order change sign of the recovery behavior parameter within the corresponding recovery process interval; the formula for calculating the equivalent process interval is... The formula for calculating the cumulative change amplitude ratio is as follows: in, They represent the first The and the first One recovery behavior parameter, This refers to the position of the corresponding parameter on the equivalent recovery process timeline. The cumulative change magnitude ratio of adjacent recovery behavior parameters; The monotonic change direction identifier, cumulative change amplitude ratio, and equivalent process interval are used as correlation elements to fill the corresponding parameter correlation matrix unit, which is used to represent the structural constraint relationship between adjacent behavioral parameters during the high-pressure slurry valve seal recovery process.

8. The method for diagnosing internal leakage in a high-pressure slurry valve according to claim 6, characterized in that: The quantitative representation of the behavioral offset includes: On the equivalent recovery process timeline, the complete recovery process is divided into several consecutive process sub-intervals; For each process sub-interval, the interval change of each recovery behavior parameter in the reference recovery behavior parameter set within that sub-interval is calculated, and the sign of the interval change is used as the reference change direction identifier for the corresponding recovery behavior parameter. For the second set of recovery behavior parameters in the current running state, the same process sub-interval division method is used to calculate the interval change of the corresponding recovery behavior parameter in each process sub-interval, and the sign of the interval change is used as the actual change direction identifier. When the actual change direction identifier of the same recovery behavior parameter is inconsistent with the reference change direction identifier within the same process sub-interval, it is determined that the recovery behavior parameter has deviated in direction within the process sub-interval. Within each process sub-interval, the number of recovery behavior parameters that deviate in direction is counted, and the ratio of the number to the total number of recovery behavior parameters included in the count within that sub-interval is defined as the direction deviation density of that process sub-interval. Based on the positional order of each process sub-interval on the equivalent recovery process time axis, a position weight function is introduced. By weighted summing of the deviation densities in each direction, the behavioral offset of the valve's current operating state relative to the reference state is obtained. The expression is: in, Indicates the number of associated terms that violate structural constraints; Indicates the first The direction of change for each associated item deviates from the identifier; These represent the ratios of the cumulative change magnitudes in the current state and in the reference state, respectively. These represent the equivalent process intervals in the current state and the reference state, respectively. The position weighting function is used to balance the contribution of different deviations. It changes monotonically as the equivalent recovery process progresses along the time axis, and is used to reduce the contribution of violation associations to the behavioral offset in the early stage of recovery and enhance the contribution of violation associations to the behavioral offset in the later stage of recovery.

9. The method for diagnosing internal leakage in a high-pressure slurry valve according to claim 1, characterized in that: The determination of the adaptive sealing state of the high-pressure slurry valve based on the change characteristics of behavioral offset includes: During multiple consecutive operation condition modulation processes, the corresponding behavior offset sequence is obtained. ,in Indicates the first The recovery process after the initial operational disturbance; Calculate the evolutionary increment between adjacent behavior offsets based on the behavior offset sequence. And construct the evolution trajectory of the behavioral offset; The evolution trajectory is segmented for consistency analysis to identify evolutionary segments in which the behavioral offset simultaneously satisfies a monotonically increasing relationship and the evolutionary increment does not decline within a preset number of consecutive perturbations. When the length of the evolution segment exceeds the preset minimum evolution length, and the behavioral offset within the evolution segment does not return to the stable interval corresponding to the sealing adaptive reference behavior model, the high-pressure slurry valve is determined to be in an internal leakage associated state.

10. The method for diagnosing internal leakage in a high-pressure slurry valve according to claim 1, characterized in that: The determination of the behavioral offset change characteristics is accomplished by performing irreversible evolutionary analysis on the behavioral offset sequence, which includes: For continuously obtained behavioral offset sequences The system is divided into segments according to a preset sliding length, and the starting offset of each segment is determined. Section termination behavior offset and the minimum behavior offset within the segment ; Based on the distribution of behavioral offsets within the segment, calculate the irreversible evolution discriminant corresponding to the segment. The calculation formula is: in, This indicates the behavioral offset corresponding to the initial disturbance of the segment; This indicates the behavioral offset corresponding to the segment termination disturbance; This represents the minimum behavioral offset that occurs within the specified segment; When the irreversible evolution discriminant When the behavior offset exceeds the preset irreversible judgment threshold in multiple consecutive segments, and the behavior offset in the corresponding segment does not return to the allowable regression range defined by the sealed adaptive reference behavior model, the behavior offset change characteristics exhibit irreversible evolution characteristics.

Images