Valve assembly quality detection method, system and device based on friction characteristics

Abstract

The application discloses a valve assembly quality detection method, system and equipment based on friction characteristics, and belongs to the technical field of intelligent manufacturing and industrial equipment quality detection. The method obtains driving load data, valve displacement data and time sequence data in the opening and closing process of a valve to be detected, and divides multiple stage detection intervals; friction characteristic parameters of each stage are extracted to construct a segmented friction characteristic sequence; the segmented friction characteristic sequence is matched with a standard friction characteristic sequence to determine friction abnormal states of different opening and closing stages, and a valve assembly quality state is determined according to the friction abnormal states. The application can realize friction abnormal stage positioning, improve the accuracy of valve assembly quality detection, and realize assembly defect identification and assembly quality evaluation.

Description

Technical Field

[0001] This invention belongs to the field of intelligent manufacturing and industrial equipment quality inspection technology, and particularly relates to valve assembly quality inspection methods, systems and equipment based on friction characteristics. Background Technology

[0002] With the development of intelligent manufacturing technology, valve production processes are gradually moving towards digitalization, automation, and intelligence. The quality of valve assembly directly affects the sealing performance, opening and closing reliability, and operational safety of fluid control systems. In current production processes, the quality of assembled valves is typically inspected through visual checks, dimensional measurements, overall opening and closing torque tests, or sealing performance tests.

[0003] Current whole-machine opening and closing tests typically evaluate a single opening and closing process as a whole, judging valve qualification based on maximum drive load, opening and closing time, or sealing results. However, valves exhibit different normal resistance characteristics during the stages of start-up disengagement, stable movement, sealing contact, and terminal clamping. Using a uniform threshold makes it difficult to distinguish between initial static friction, kinematic friction, sealing contact resistance, and terminal clamping resistance. When valve stem eccentricity, abnormal guide fit, seal installation deviation, or abnormal preload leads to an increase in the overall drive load, existing methods struggle to determine the specific stage of the abnormality and the corresponding assembly defects.

[0004] Therefore, existing valve assembly quality inspection methods suffer from insufficient process state perception, difficulty in locating stage friction anomalies, and limited information dimensions in the inspection results. There is an urgent need for an inspection technology that can perform staged inspections of the valve opening and closing process and determine the type of assembly defect, anomaly level, and assembly quality status based on the friction characteristics of different stages. Summary of the Invention

[0005] This invention provides a valve assembly quality inspection method, system, and equipment based on friction characteristics. It aims to solve the technical problems of existing valve assembly quality inspection methods that evaluate the entire opening and closing process as a whole, making it difficult to distinguish the normal resistance characteristics of different opening and closing stages, and making it difficult to locate abnormal friction stages, identify assembly defect types, and determine the abnormality level based solely on the maximum driving load.

[0006] Firstly, the present invention is implemented as follows: a valve assembly quality inspection method based on friction characteristics, comprising: The driving load data, valve displacement data and corresponding time series data of the valve to be tested during the opening and closing process are obtained, and the opening and closing process is divided into multiple detection intervals based on the valve displacement data. For each of the aforementioned detection intervals, the friction characteristic parameters of the corresponding stage are determined based on the relationship between the driving load data and the valve displacement data, and a segmented friction characteristic sequence of the valve to be tested is constructed based on the friction characteristic parameters corresponding to different detection intervals. The segmented friction feature sequence is matched and analyzed with the standard friction feature sequence corresponding to the standard valve to determine the abnormal friction state of the valve under test in different opening and closing stages. The assembly quality status of the valve to be tested is determined based on the abnormal friction states during different opening and closing stages.

[0007] Preferably, the valve displacement data divides a single opening and closing process into multiple detection intervals, including: The valve displacement change rate is determined based on the valve displacement data and the corresponding time series data, and the drive load change rate is determined based on the drive load data. Based on the coordinated relationship between the valve displacement change rate and the drive load change rate, identify the nodes of motion state change during the opening and closing process corresponding to static start-up, stable motion, sealing contact, and terminal pressing. According to the displacement position corresponding to each of the motion state change nodes, a single opening and closing process is divided into the starting disengagement stage, the stable motion stage, the sealing contact stage, and the terminal pressing stage. The start-up disengagement phase corresponds to the process by which the valve under test overcomes the initial static frictional resistance and begins to move; the stable movement phase corresponds to the frictional change process during the continuous movement of the valve under test; the sealing contact phase corresponds to the resistance change process after the sealing pair begins to contact; and the terminal pressing phase corresponds to the resistance change process after the sealing pair forms a pre-tightening force.

[0008] Preferably, determining the friction characteristic parameters for the corresponding stage based on the relationship between the drive load data and the valve displacement data includes: Acquire drive load data, valve displacement data, and time series data within the corresponding detection interval; Based on the drive load data and the valve displacement data, a load response curve of the drive load as the valve displacement changes is established; At least one of the following parameters is extracted from the load response curve: average load value, peak load value, load fluctuation amplitude, load change rate, and load change per unit displacement, as the friction characteristic parameter for the corresponding stage. Friction feature vectors for the corresponding detection interval are formed based on the friction feature parameters of the corresponding stage.

[0009] Preferably, a segmented friction feature sequence of the valve to be tested is constructed based on the friction feature parameters corresponding to different detection intervals, including: Obtain the friction feature vectors corresponding to each of the detection intervals in each stage; Based on the displacement stroke, stage duration, and driving load reference value of the corresponding stage detection interval, the friction characteristic parameters in the friction characteristic vector of each stage are normalized without dimension. According to the sequence of each stage of the opening and closing process, the normalized friction feature vectors of each stage are combined sequentially to form a segmented friction feature sequence corresponding to one opening and closing process.

[0010] Preferably, before performing matching analysis between the segmented friction feature sequence and the standard friction feature sequence corresponding to the standard valve, the method further includes: Acquire the drive load data, valve displacement data, and time series data of multiple qualified valves of the same model during the opening and closing process under the same testing conditions; For each valve that meets the assembly quality standards, the stage inspection interval is divided, friction feature parameters are extracted, and stage normalization is performed to obtain the segmented friction feature sequence corresponding to each valve that meets the assembly quality standards. Statistical analysis is performed on the multiple segmented friction feature sequences according to their corresponding opening and closing stages to determine the standard friction feature vector corresponding to each opening and closing stage. The standard friction feature vectors are combined according to the order of each stage of the opening and closing process to establish a standard friction feature sequence corresponding to the same type of valve, and the standard friction feature sequence is stored in the standard feature database.

[0011] Preferably, the segmented friction feature sequence is matched and analyzed with the standard friction feature sequence corresponding to the standard valve to determine the abnormal friction state of the valve under test in different opening and closing stages, including: The segmented friction feature sequence corresponding to the valve to be tested is matched with the standard friction feature sequence according to the opening and closing stages; Determine the degree of feature deviation between the friction feature vector of the valve to be tested and the corresponding standard friction feature vector in each opening and closing stage; The degree of deviation of the feature is compared with the anomaly determination threshold of the corresponding opening and closing stage; When the degree of deviation of the feature is greater than the abnormal judgment threshold of the corresponding opening and closing stage, it is determined that there is an abnormal friction state in the corresponding opening and closing stage; otherwise, it is determined that the corresponding opening and closing stage is a normal friction state. Based on the abnormal friction states corresponding to each opening and closing stage, the stage friction anomaly distribution results of the valve to be tested are formed.

[0012] Preferably, determining the assembly quality status of the valve to be tested based on the abnormal friction states during different opening and closing stages includes: The abnormal stage is determined based on the abnormal friction state corresponding to each opening and closing stage. Based on the pre-established correspondence between the opening and closing stages and the assembly defect types, the assembly defect type corresponding to each abnormal stage is determined. Based on the assembly defect type and the degree of deviation of the corresponding opening and closing stage, the abnormality level of each assembly defect is determined. Based on the abnormality level and distribution of each assembly defect, the assembly quality status of the valve to be tested is determined, and the assembly quality test results are output.

[0013] Secondly, this application discloses a valve assembly quality inspection system based on friction characteristics, including: A load sensor is driven and installed at the drive mechanism of the valve to be tested; A displacement sensor is installed on the moving part of the valve to be detected; The controller includes a processor and a memory connected to the processor, the processor being electrically connected to the drive load sensor and the displacement sensor respectively; The detection result output device is connected to the processor; The memory stores a standard friction feature sequence and a computer program corresponding to a standard valve. When the processor executes the computer program, it implements the valve assembly quality detection method based on friction features as described in the first aspect and sends the assembly quality detection result to the detection result output device.

[0014] Thirdly, the present invention discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the valve assembly quality inspection method based on friction characteristics as described in the first aspect. Fourthly, the present invention discloses a computer device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor; When the processor executes the computer program, it implements the valve assembly quality inspection method based on friction characteristics as described in the first aspect.

[0015] Compared with the prior art, the present invention has the following beneficial effects: This invention divides a single opening and closing process into multiple detection stages based on valve displacement data, and detects the starting disengagement stage, the stable movement stage, the sealing contact stage, and the terminal pressing stage respectively. This allows the overall opening and closing load to be decomposed into different movement stages, enabling stage-based location of friction abnormalities.

[0016] This invention extracts friction feature parameters for different stages and performs dimensionless normalization based on displacement stroke, stage duration, and driving load reference value. This reduces the impact of differences in stroke, duration, and load magnitude between different stages on feature comparison and reduces misjudgments caused by using a uniform threshold. Attached Figure Description

[0017] Figure 1 This is a flowchart of the valve assembly quality inspection method based on friction characteristics according to the present invention.

[0018] Figure 2 This is a flowchart illustrating the detection interval division during the opening and closing process of this invention.

[0019] Figure 3 This is a flowchart of the friction feature parameter extraction and segmented friction feature sequence construction of the present invention.

[0020] Figure 4 A flowchart for establishing the standard friction feature sequence of this invention.

[0021] Figure 5 This is a flowchart illustrating the process of matching and determining abnormal friction states during the stages of this invention.

[0022] Figure 6 This is a flowchart for determining the assembly defect type and assembly quality status of the present invention.

[0023] Figure 7 This is a schematic diagram of the valve assembly quality inspection system based on friction characteristics according to the present invention. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0025] This embodiment uses an electrically driven linear stop valve as an example. The valve to be tested includes a valve body, valve seat, valve core, valve stem, guide element, seal, and electric drive mechanism. The electric drive mechanism drives the valve stem to move linearly along the valve axis through a transmission mechanism. The valve core is connected to the valve stem and moves synchronously with the valve stem. When the valve is closed, the valve core gradually approaches and contacts the valve seat, subsequently forming a sealing preload. When the valve is opened, the valve core disengages from the valve seat and moves away from the valve seat.

[0026] In other embodiments, the valve to be detected may also be a gate valve, ball valve, butterfly valve, regulating valve, or other valves with opening and closing moving parts. For a gate valve, the moving part may be a valve stem, a gate, or a transmission part that moves synchronously with it; for a ball valve, the moving part may be a valve stem, a valve ball, or a rotary transmission shaft; for a butterfly valve, the moving part may be a valve shaft, a valve plate, or a transmission part that moves synchronously with it.

[0027] like Figure 1 As shown, the detection method in this embodiment includes S100, S200, S300 and S400.

[0028] Step S100: Obtain the drive load data, valve displacement data and corresponding time series data during the opening and closing process of the valve to be tested, and divide one opening and closing process into multiple stage detection intervals based on the valve displacement data.

[0029] Before the valve to be tested performs an opening and closing action, the drive load sensor is set at the output shaft, transmission shaft, coupling, push rod, or drive motor power supply circuit of the electric drive mechanism, and the displacement sensor is set at the valve stem, valve core, or transmission component that moves synchronously with the valve stem.

[0030] Drive load sensors are used to acquire drive load data. Drive load data can be drive torque, drive thrust, drive current, hydraulic pressure, or pneumatic pressure. When a torque sensor or thrust sensor is used, the drive load data directly characterizes the mechanical load required for the drive mechanism to overcome the mechanical resistance of the valve. When drive current is used, it can be converted into the corresponding drive torque or drive thrust based on the torque coefficient, transmission ratio, and transmission efficiency of the drive motor. When hydraulic pressure or pneumatic pressure is used, the pressure can be converted into the drive thrust or drive torque acting on the valve stem or valve shaft based on the effective area of ​​the actuator piston, mechanical transmission ratio, and transmission efficiency.

[0031] Displacement sensors are used to acquire valve displacement data. For linear valves, magnetostrictive displacement sensors, draw-wire displacement sensors, grating rulers, or linear encoders can be used to acquire the axial displacement of the valve stem or valve core; for angular valves, rotary encoders, angular displacement sensors, or absolute encoders can be used to acquire the rotation angle of the valve shaft or valve plate, and the rotation angle is used as the valve displacement data.

[0032] The controller generates a unified sampling clock and synchronously collects drive load data and valve displacement data according to the same sampling period. Each sampling point is appended with a corresponding timestamp to form time-series data. The sampling frequency can be set from 50Hz to 2000Hz depending on the valve opening and closing speed and detection accuracy. This embodiment uses 500Hz, that is, drive load data and valve displacement data are collected once every 2ms.

[0033] For data obtained from different acquisition interfaces with inconsistent sampling times, time alignment is performed according to a unified time axis. Specifically, the sampling time of the displacement data can be selected as the target time point, and linear interpolation, cubic spline interpolation, or hold-behind interpolation can be used to obtain the drive load data corresponding to each target time point. In this embodiment, linear interpolation is used so that each drive load data has the corresponding valve displacement data at the same timestamp.

[0034] During data preprocessing, median filtering can be used to remove isolated pulse values, and moving average filtering or low-pass filtering can be used to reduce high-frequency noise. Sampling points exceeding the sensor's range, differing from previous and subsequent samples by more than a preset multiple, or continuously missing are marked as outliers. Single or a small number of outliers can be replaced by interpolation of adjacent valid data; when the number of consecutive outliers exceeds a preset limit, the current detection data is deemed invalid, and the start / stop detection is re-executed.

[0035] like Figure 2 As shown, the process of dividing the detection interval into stages includes S110, S120 and S130.

[0036] S110, determine the valve displacement change rate based on the valve displacement data and the corresponding time series data, and determine the drive load change rate based on the drive load data.

[0037] Let the valve displacement data at the i-th sampling point be x. i The corresponding time is t i Then the valve displacement change rate v i It can be determined according to the following formula:

[0038] Among them, v i x represents the rate of change of valve displacement corresponding to the i-th sampling point; i and x i－1 These represent the valve displacement data at two adjacent sampling points; t i and t i－1 These represent the times corresponding to two adjacent sampling points.

[0039] To reduce the impact of single-point noise, a sliding time window of length m can also be used to determine the valve displacement change rate.

[0040] Let the driving load data of the i-th sampling point be F. i Then the rate of change of the driving load r i It can be determined according to the following formula:

[0041] Where, ri F represents the rate of change of the driving load corresponding to the i-th sampling point; i and F i－1 These represent the driving load data of two adjacent sampling points.

[0042] The valve displacement rate and the driving load rate are aligned using the same timestamp. When the two are processed through filtering windows of different lengths, the filtered results are mapped to a unified time axis, and a pair of co-changing data is formed using vi and ri at the same sampling time.

[0043] S120, based on the coordinated change relationship between the valve displacement change rate and the drive load change rate, identify the motion state change nodes corresponding to static start-up, stable movement, sealing contact, and terminal pressing during the opening and closing process.

[0044] The controller identifies the motion state change nodes corresponding to static start-up, stable movement, sealing contact, and terminal clamping based on the coordinated change relationship between the valve displacement change rate and the drive load change rate.

[0045] During the valve closing process, when the valve displacement change rate changes from near zero to greater than the start-up speed threshold, and the drive load change rate shows a positive peak within the corresponding time window, this moment is defined as the static start-up node. The displacement position corresponding to the static start-up node is recorded as the first displacement position.

[0046] When the valve displacement rate remains within a stable speed range for a continuous preset time, and the absolute value of the driving load change rate decreases to within the stable change rate threshold, this moment is defined as the stable motion node. The displacement position corresponding to the stable motion node is recorded as the second displacement position.

[0047] As the valve core approaches the valve seat, the valve displacement rate begins to decrease, while the driving load rate continues to increase. When the decrease in the valve displacement rate exceeds the contact velocity change threshold, and the driving load rate changes at multiple consecutive sampling points are greater than the contact load rate change threshold, this moment is defined as the sealing contact node. The displacement position corresponding to the sealing contact node is recorded as the third displacement position.

[0048] When the valve displacement rate decreases further and approaches zero, while the drive load data continues to increase or remains in the high load range, this moment is defined as the final clamping point. The displacement position corresponding to the final clamping point is recorded as the fourth displacement position.

[0049] The aforementioned thresholds can be determined using statistical data from multiple qualified valves of the same model under the same testing conditions. For example, the mean and standard deviation of the valve displacement change rate and the driving load change rate near each node of a qualified valve can be used as a basis, and the corresponding thresholds can be set by adding or subtracting a preset multiple of the standard deviation from the mean.

[0050] During valve opening, the physical processes of each stage are the reverse of those during valve closing. The valve first overcomes the pre-tightening resistance formed by the terminal clamping, then the sealing pair disengages, enters a stable motion, and finally reaches the opening endpoint. To maintain consistency in stage terminology, the intervals corresponding to the terminal clamping release stage, the sealing contact disengagement stage, the stable motion stage, and the start-up disengagement stage can be determined sequentially according to the actual motion state, and the opening and closing direction identifiers can be set in the data structure. Alternatively, segmented friction characteristic sequences for the valve opening and closing processes can be established separately.

[0051] For ball valves or butterfly valves, the valve displacement data is the rotation angle. The sealing contact node corresponds to the position where the valve ball or valve plate and the sealing ring begin to form significant contact resistance. The terminal pressing node corresponds to the position where the actuator further drives the valve ball or valve plate to press the sealing ring or reach the mechanical limit.

[0052] S130, according to the displacement position corresponding to each of the motion state change nodes, the opening and closing process is divided into the starting disengagement stage, the stable motion stage, the sealing contact stage and the terminal pressing stage.

[0053] Based on the displacement position corresponding to each node of motion change, the valve closing process is divided into the start-up disengagement stage, the stable motion stage, the sealing contact stage, and the terminal pressing stage.

[0054] The start-up disengagement stage is the process by which the valve under test starts to move from a stationary state and overcomes the initial static frictional resistance; the stable motion stage is the process of frictional changes during the continuous movement of the valve's moving parts; the sealing contact stage is the process of resistance changes after the sealing pair begins to contact; and the final tightening stage is the process of resistance changes after the sealing pair forms a pre-tightening force.

[0055] In this embodiment, the period between the first displacement position and the second displacement position forms a start-up disengagement stage, the period between the second displacement position and the third displacement position forms a stable movement stage, the period between the third displacement position and the fourth displacement position forms a sealing contact stage, and the period between the fourth displacement position and the valve closing termination position forms a terminal pressing stage.

[0056] Step S200: For each stage detection interval, determine the friction characteristic parameters of the corresponding stage based on the change relationship between the drive load data and the valve displacement data, and construct the segmented friction characteristic sequence of the valve to be detected based on the friction characteristic parameters corresponding to different stage detection intervals.

[0057] like Figure 3 As shown, the process of extracting friction feature parameters includes S210, S220, S230 and S240.

[0058] S210, acquire the drive load data, valve displacement data and time series data within the corresponding stage detection interval.

[0059] The controller extracts the drive load data, valve displacement data, and time series data within the corresponding interval from the synchronously collected data according to the start and end times of each detection interval.

[0060] S220, Based on the drive load data and the valve displacement data, establish a load response curve of the drive load as the valve displacement changes.

[0061] Using valve displacement data as the x-axis and drive load data as the y-axis, a load response curve is established to show the change in drive load with valve displacement. For linear valves, the x-axis unit can be millimeters, and the y-axis unit can be Newtons, Newton-meters, or Ampers; for angular valves, the x-axis unit can be degrees or radians.

[0062] When multiple drive load samples correspond to the same displacement position, the data can be binned according to displacement intervals, and the mean or median of the drive load data within each displacement interval can be used as the drive load value corresponding to that displacement position. Alternatively, the drive load data can be interpolated according to a uniform displacement interval to form a load response curve with equal displacement intervals.

[0063] S230, extract at least one of the following from the load response curve: average load value, peak load value, load fluctuation amplitude, load change rate, and load change per unit displacement, as friction characteristic parameters for the corresponding stage.

[0064] Suppose there are n sampling points in a certain detection interval, and the corresponding driving load data are F1, F2, ..., F n .

[0065] The average load value can be determined by the following formula:

[0066] Peak load can be determined using the following formula:

[0067] The load fluctuation amplitude can be determined by the difference between the maximum and minimum values:

[0068] Among them, F min This represents the minimum drive load data within this phase.

[0069] The load change rate can be represented by the average absolute value, maximum value, or fitted slope of the first-order rate of change of the driving load data with respect to time during that period. In this embodiment, the linear fitted slope of the driving load changing with time is used as the load change rate.

[0070] The change in load per unit displacement can be determined by the following formula:

[0071] Among them, F start and F end These are the drive load data corresponding to the start and end positions of this stage, x start and x end These are the initial and final displacements for this stage, respectively.

[0072] In other embodiments, the unit displacement load change within each displacement interval can be determined by locally fitting the load response curve, and its average value, maximum value or quantile can be used as the friction characteristic parameter of that stage.

[0073] S240, based on the friction characteristic parameters of the corresponding stage, forms the friction feature vector of the detection interval of the corresponding stage.

[0074] In this embodiment, a five-dimensional friction feature vector is used for each stage:

[0075] Where j represents the start-up and stop-down phases; F avg This represents the average load value; F max This represents the peak load value; F a R represents the load fluctuation amplitude. t K represents the load change rate; x This represents the change in load per unit displacement.

[0076] The parameters are arranged in a fixed order: average load value, peak load value, load fluctuation amplitude, load change rate, and load change per unit displacement. The same parameter arrangement order is used in different stages to facilitate corresponding matching.

[0077] The process of constructing the segmented friction feature sequence includes S250, S260, and S270.

[0078] S250, respectively obtain the friction feature vector corresponding to each of the detection intervals of the aforementioned stages.

[0079] The friction feature vectors corresponding to the start-up separation stage, stable motion stage, sealing contact stage, and terminal pressing stage are obtained respectively and denoted as P1, P2, P3, and P4.

[0080] S260, based on the displacement stroke, stage duration, and driving load reference value of the corresponding stage detection interval, performs dimensionless normalization processing on the friction characteristic parameters in the friction characteristic vector of each stage.

[0081] The displacement stroke at each stage is determined based on the difference between the ending displacement and the starting displacement of the corresponding stage:

[0082] Among them, L j This represents the displacement distance in stage j.

[0083] The duration of a phase is determined by the difference between the end time and the start time of the corresponding phase:

[0084] Among them, T j This indicates the duration of stage j.

[0085] The drive load reference value can be the rated drive load of a standard valve of the same model under the same testing conditions, the average load value of multiple qualified valves in a given stage, or the rated output load of the corresponding drive mechanism. In this embodiment, the average load of multiple qualified valves in the corresponding stage is used as the drive load reference value F. bj .

[0086] For the average load value, peak load value, and load fluctuation amplitude, they can be normalized by dividing by the corresponding drive load reference value:

[0087] For the load change rate, it can be normalized based on the driving load baseline value and the stage duration:

[0088] The change in load per unit displacement can be normalized based on the driving load reference value and the displacement stroke:

[0089] The normalized parameters are dimensionless. Under normal circumstances, each parameter can be mapped to a range of 0 to 1 or other preset values ​​through statistical range or limiting processing. For example, when the normalization result exceeds a preset upper limit, it can be limited to the preset upper limit; or the parameter can be mapped to a range of 0 to 1 using a minimum-maximum value normalization method.

[0090] S270, according to the order of each stage of the opening and closing process, the normalized friction feature vectors of each stage are sequentially combined to form a segmented friction feature sequence corresponding to one opening and closing process.

[0091] According to the chronological order of the opening and closing processes, the normalized friction feature vectors of each stage are combined sequentially:

[0092] Where Q represents the segmented friction feature sequence corresponding to one opening and closing process; P′1, P′2, P′3 and P′4 represent the normalized friction feature vectors corresponding to the starting disengagement stage, the stable motion stage, the sealing contact stage and the terminal pressing stage, respectively.

[0093] When detecting the valve opening and closing processes separately, separate segmented friction feature sequences for valve opening and valve closing can be established and matched with standard friction feature sequences for the corresponding directions. Alternatively, the segmented friction feature sequences for valve opening and valve closing can be sequentially combined to form a joint segmented friction feature sequence that includes both the valve opening and closing processes, with an opening / closing direction field set in the sequence.

[0094] Step S300: Match the segmented friction feature sequence with the standard friction feature sequence corresponding to the standard valve to determine the abnormal friction state of the valve under test in different opening and closing stages.

[0095] like Figure 4 As shown, the process of establishing the standard friction feature sequence includes steps S310 to S340.

[0096] S310: Obtain the drive load data, valve displacement data, and time series data of multiple valves of the same model that have passed assembly quality and are performing opening and closing actions under the same testing conditions.

[0097] Multiple valves of the same model that have passed assembly quality tests are selected as standard valves. Valves that pass assembly quality tests should undergo visual inspection, critical dimension inspection, sealing performance testing, rated operating condition opening and closing tests, and necessary disassembly and inspection for verification. For samples used to establish a standard feature database, valve stem coaxiality, guide clearance, seal installation position, and terminal preload may also be required to meet the requirements of the corresponding product's technical documents.

[0098] Valves of the same model should have at least the same valve type, nominal diameter, nominal pressure, valve body structure, valve core or valve plate structure, sealing specifications, drive mechanism model and transmission ratio.

[0099] The same testing conditions should include at least the same or allowable deviation range of drive method, opening and closing speed, medium state, and environmental conditions. The medium state can include no medium, air, water, or other specified media, as well as the corresponding pressure and temperature; the environmental conditions can include ambient temperature, ambient humidity, and power supply voltage.

[0100] The number of qualified valve samples can be set to more than 10 depending on the production batch and statistical requirements. In this example, 30 qualified valves of the same model are used.

[0101] S320 performs stage inspection interval division, friction feature parameter extraction and stage normalization processing on each qualified assembly valve to obtain the segmented friction feature sequence corresponding to each qualified assembly valve.

[0102] For each qualified valve, the process involves dividing the testing interval into stages, extracting friction feature parameters, and performing dimensionless normalization to form a corresponding segmented friction feature sequence.

[0103] Before statistical analysis, outlier samples can be removed from the friction characteristic parameters at each stage. Outlier removal can be performed using the interquartile range method, the three-standard-deviation method, or a method based on the absolute deviation of the median. For example, when a certain characteristic parameter deviates from the sample median by more than three times the absolute deviation of the median, the corresponding sample is marked as an outlier, and it is checked whether there are any abnormalities in the detection conditions or sensor malfunctions.

[0104] S330, statistical analysis is performed on the multiple segmented friction feature sequences according to their corresponding opening and closing stages to determine the standard friction feature vector corresponding to each opening and closing stage.

[0105] Statistical analysis was performed on multiple segmented friction feature sequences according to their corresponding opening and closing stages. For each opening and closing stage and each friction feature parameter, the mean after removing outliers could be used as the standard value, or the median, truncated mean, or preset quantile could be used as the standard value.

[0106] In this embodiment, the mean value after removing abnormal samples is used to form the standard friction feature vector corresponding to each opening and closing stage, and the standard deviation, allowable range and abnormal judgment threshold of each parameter are stored simultaneously.

[0107] S340, combine the standard friction feature vectors according to the order of each stage of the opening and closing process, establish a standard friction feature sequence corresponding to the same model valve, and store the standard friction feature sequence in the standard feature database.

[0108] The standard friction feature vectors are combined in the order of the start-up separation stage, the stable motion stage, the sealing contact stage, and the terminal pressing stage to form a standard friction feature sequence.

[0109] The standard feature database can be a relational database, a key-value database, or a local structured file. The database can include indexes for valve model, testing conditions, opening / closing direction, and version. Each record must include at least the valve model, actuation method, opening / closing speed, media state, environmental conditions, stage identifier, standard friction feature vector, parameter standard deviation, anomaly detection threshold, number of samples, and update time.

[0110] When the valve structure, materials, sealing specifications, drive mechanism model, or assembly process changes, or when the number of newly qualified samples reaches a preset quantity, the standard friction feature vector can be recalculated and the standard friction feature sequence updated. Historical versions are retained during the update process for traceability of test results.

[0111] like Figure 5 As shown, the stage matching and friction anomaly identification include S350 to S390.

[0112] S350, the segmented friction feature sequence corresponding to the valve to be tested is matched with the standard friction feature sequence according to the opening and closing stages.

[0113] Based on the valve model, testing conditions, and opening / closing direction, the corresponding standard friction feature sequence is retrieved from the standard feature database. The valve's start-up disengagement stage, stable movement stage, sealing contact stage, and terminal clamping stage are matched one by one with the corresponding stages in the standard friction feature sequence.

[0114] S360, determine the degree of feature deviation between the friction feature vector corresponding to the valve to be tested and the corresponding standard friction feature vector in each opening and closing stage.

[0115] Let the normalized friction feature vector of the valve to be tested in the j-th stage be:

[0116] The corresponding standard friction feature vector is:

[0117] The degree of feature deviation Dj in stage j can be determined according to the weighted distance:

[0118] Among them, w k The weight of the k-th friction feature parameter is represented by ε, and the sum of all weights is 1; ε is a positive number set to prevent the denominator from being zero; k represents the number of friction feature parameters.

[0119] The weights of different friction characteristic parameters can be determined based on the ability to identify historical defect samples, expert experience, or statistical analysis. For example, the weights of peak load and load change rate can be increased during the start-up disengagement phase; the weights of average load and load fluctuation amplitude can be increased during the stable motion phase; and the weights of load change per unit displacement and peak load can be increased during the sealing contact phase and the terminal clamping phase.

[0120] S370, compare the degree of deviation of the feature with the anomaly determination threshold of the corresponding opening and closing stage.

[0121] Different abnormal judgment thresholds are set for different opening and closing stages. The reason is that the starting and disengaging stage is mainly affected by initial static friction, the stable movement stage is mainly affected by motion friction and guide fit, and the sealing contact stage and the terminal pressing stage are mainly affected by the sealing pair contact and pre-tightening state. The normal load characteristics and allowable deviations of each stage are different, so a uniform threshold cannot be used.

[0122] The anomaly detection threshold can be determined based on the statistical distribution of the deviations in characteristics of multiple qualified valves at corresponding stages. For example, the mean of the deviations in the qualified sample stages plus three standard deviations can be used as the anomaly detection threshold, or the 99th percentile of the deviations in the qualified sample stages can be used as the anomaly detection threshold.

[0123] S380, when the degree of deviation of the feature is greater than the abnormal judgment threshold of the corresponding opening and closing stage, it is determined that there is an abnormal friction state in the corresponding opening and closing stage; otherwise, it is determined that the corresponding opening and closing stage is a normal friction state.

[0124] When the characteristic deviation degree D in stage j j The anomaly detection threshold H is greater than the corresponding stage. j When D is present, it is determined that an abnormal frictional state exists at this stage; when D j Not greater than H j At that time, this stage is determined to be the normal friction state.

[0125] S390, Based on the frictional abnormality state corresponding to each opening and closing stage, form the stage frictional abnormality distribution result of the valve to be tested.

[0126] The results of stage friction anomaly distribution can be represented by arrays, structures, tables, or data messages. Each stage corresponds to one record, and the record content includes at least the stage identifier, the degree of feature deviation, the anomaly judgment threshold, the friction state, and the friction feature parameter with the largest deviation.

[0127] For example, the results of stage friction anomaly distribution can be expressed as: Initiation and disengagement phase: Normal friction state; Stable motion phase: abnormal friction state; Sealed contact stage: Normal friction state; Terminal pressing stage: abnormal friction state.

[0128] Taking the five-dimensional normalized friction feature vector as an example, the friction feature vector of a valve under test during its stable motion phase is as follows: [0.82, 0.86, 0.31, 0.18, 0.22]; The corresponding standard friction feature vector is: [0.80, 0.84, 0.30, 0.17, 0.21].

[0129] With each parameter having a weight of 0.2, the calculated feature deviation is less than the anomaly detection threshold for this stage, thus determining the stable motion stage as a normal friction state.

[0130] The friction feature vector of a certain valve under test during the terminal clamping stage is: [1.18, 1.35, 0.42, 0.68, 1.26]; The corresponding standard friction feature vector is: [1.00, 1.05, 0.28, 0.40, 0.90].

[0131] After calculation according to the preset weights, the degree of feature deviation is greater than the abnormal judgment threshold of the terminal pressing stage, so it is determined that there is an abnormal friction state in the terminal pressing stage.

[0132] Step S400: Determine the assembly quality status of the valve to be tested based on the abnormal friction status during different opening and closing stages.

[0133] like Figure 6 As shown, the assembly quality status determination process includes S410 to S440.

[0134] S410, determine the abnormal stage based on the abnormal friction state corresponding to each opening and closing stage.

[0135] Based on the results of the abnormal distribution of friction in each stage, the opening and closing stages that are in an abnormal friction state are identified as abnormal stages.

[0136] S420, based on the pre-established correspondence between the opening and closing stages and the assembly defect types, determine the assembly defect type corresponding to each abnormal stage.

[0137] Establish a correspondence between the opening and closing stages and the types of assembly defects in advance.

[0138] Abnormalities during the start-up disengagement phase may be due to valve stem misalignment, excessively tight guide fit, excessive initial preload, or insufficient lubrication.

[0139] Abnormalities during the stable motion phase can correspond to abnormal coaxiality between the valve stem and guide, abnormal clearance of the moving parts, local jamming, or abnormal lubrication.

[0140] Abnormalities in the sealing contact stage may correspond to misaligned installation of the seal, misalignment of the sealing pair, foreign objects on the sealing surface, or dimensional deviations of the seal.

[0141] Abnormalities in the terminal tightening stage may correspond to insufficient sealing pre-tightening force, excessive sealing pre-tightening force, deviation of limit position, or abnormal tightening state.

[0142] The above correspondence can be configured according to different valve types and structures, and is not intended to limit each abnormal stage to only a single assembly defect type.

[0143] When multiple candidate assembly defects correspond to the same abnormal stage, further judgment can be made based on the friction characteristic parameters with the largest deviation, the displacement location where the abnormality occurs, the symmetry of the valve opening and closing process, and the abnormal combination of adjacent stages. For example, if the peak load value increases significantly during the start-up and disengagement stage and the average load value is basically normal during the stable motion stage, it can be preemptively judged as excessive initial preload or insufficient lubrication; if the load fluctuation amplitude continues to increase during the stable motion stage, it can be preemptively judged as local jamming or coaxiality abnormality.

[0144] S430, determine the abnormality level of each assembly defect based on the type of each assembly defect and the degree of deviation of the characteristics of the corresponding opening and closing stages.

[0145] Anomalies can be classified into three levels: Level 1, Level 2, and Level 3. The anomaly level can be determined based on the ratio of the degree of feature deviation to the anomaly determination threshold.

[0146] When the deviation of a feature is greater than the anomaly determination threshold but not greater than the first-level threshold, it is determined to be a level 1 anomaly; when the deviation of a feature is greater than the first-level threshold but not greater than the second-level threshold, it is determined to be a level 2 anomaly; when the deviation of a feature is greater than the second-level threshold, it is determined to be a level 3 anomaly.

[0147] The first-level threshold and the second-level threshold can be set to 1.5 times and 2 times the anomaly judgment threshold, respectively, or they can be determined based on historical defect samples and assembly process requirements.

[0148] Different levels of anomaly can be assigned different weights. For example, the terminal clamping stage is directly related to sealing performance and can be assigned a higher weight; the stable motion stage mainly reflects the coordination state of the kinematic pairs and can be assigned a medium weight.

[0149] S440: Based on the abnormality level and distribution of each assembly defect, determine the assembly quality status of the valve to be inspected, and output the assembly quality inspection results.

[0150] Assembly quality status can be divided into qualified, awaiting re-inspection, and unqualified.

[0151] When all stages are in a normal friction state, the assembly quality of the valve to be tested is determined to be qualified.

[0152] When a Level 1 anomaly exists and no Level 2 or Level 3 anomalies exist, the assembly quality status is determined to be pending re-inspection.

[0153] When there are level 2 or level 3 anomalies, or when anomalies exist in multiple stages simultaneously, the assembly quality status is determined to be unqualified.

[0154] The assembly quality inspection results should include at least the valve identification, inspection time, abnormal stage, assembly defect type, abnormality level, and overall assembly quality status. They may also include friction feature vectors for each stage, standard friction feature vectors, feature deviation degree, and abnormality judgment threshold.

[0155] This embodiment takes the valve closure detection of an electric linear stop valve as an example.

[0156] During the detection process, the drive load sensor is a thrust sensor with a range of 0 to 20 kN; the displacement sensor is a linear displacement sensor with a range of 0 to 100 mm; the sampling frequency is 500 Hz. The controller synchronously collects drive load data, valve displacement data, and time series data, and processes them through median filtering and moving average.

[0157] Based on the coordinated relationship between the valve displacement change rate and the driving load change rate, four motion state change nodes were identified, corresponding to displacement positions of 2.0mm, 8.0mm, 72.0mm, and 88.0mm. Therefore, the following classifications are made: Initiation and disengagement phase: 2.0mm to 8.0mm; Stable motion phase: 8.0mm to 72.0mm; Sealing contact phase: 72.0 mm to 88.0 mm; Terminal clamping stage: 88.0mm to the valve closing position.

[0158] During the terminal compression phase, the following was extracted: The average load value is 8.6 kN; The peak load is 10.2 kN; The load fluctuation amplitude is 2.1kN; The load change rate is 1.6 kN / s; The unit displacement load change is 0.82 kN / mm.

[0159] The displacement stroke in this stage is 5 mm, the stage duration is 1.2 s, and the driving load reference value is 8.0 kN. After dimensionless normalization, the friction characteristic vector of the final clamping stage is obtained: [1.075, 1.275, 0.2625, 0.24, 0.5125].

[0160] The standard friction feature vector for the terminal clamping stage of the corresponding valve model in the standard feature database is as follows: [0.96, 1.08, 0.18, 0.16, 0.35].

[0161] The deviation of the feature at this stage was calculated to be 0.31 using preset weights, corresponding to an anomaly detection threshold of 0.18. Since 0.31 is greater than 0.18, it is determined that there is an abnormal friction state during the terminal pressing stage.

[0162] Further, based on the unit displacement load change and peak load value with the largest deviation, and combined with the correspondence between the opening and closing stages and assembly defect types, the candidate assembly defect types were determined to be excessive sealing preload or limit position deviation. The deviation of this stage was greater than the anomaly judgment threshold but not more than twice it, thus it was determined to be a level two anomaly. The other three stages were all under normal friction conditions. Finally, the assembly quality of the valve under test was determined to be unqualified, and the anomaly stage, candidate assembly defect type, level two anomaly, and unqualified status were output.

[0163] This completes the entire process from raw drive load data and valve displacement data to assembly quality inspection results.

[0164] like Figure 7 As shown, the valve assembly quality inspection system based on friction characteristics includes a drive load sensor, a displacement sensor, a controller, and a test result output device.

[0165] The drive load sensor is located at the drive mechanism of the valve to be tested. For electric linear valves, the drive load sensor can be located between the output shaft of the drive motor and the valve stem transmission mechanism to detect the drive torque; it can also be located on the force path of the push rod or valve stem to detect the axial drive thrust; alternatively, a current sensor can be used to detect the drive motor current, and the processor can convert it into drive load data based on the motor torque coefficient and transmission parameters.

[0166] The displacement sensor is installed at the moving part of the valve to be tested. The moving part can be the valve stem, valve core, valve plate, or a transmission component that moves synchronously with it. For linear valves, the displacement sensor can be fixed on the valve bracket or testing fixture, and its testing end is connected to the valve stem or push rod; for rotary valves, the angular displacement sensor or encoder can be installed at the valve shaft or the output shaft of the actuator.

[0167] The controller includes a processor, memory, a data acquisition interface, and a communication interface. The processor is electrically connected to the drive load sensor and displacement sensor via the data acquisition interface. The data acquisition interface may include an analog signal acquisition circuit, an analog-to-digital converter, a digital input interface, an encoder interface, or a fieldbus interface.

[0168] The sensor and processor can be connected via wired or wireless means. Wired connections can use shielded cables, industrial Ethernet, RS485, or CAN bus; wireless connections can use industrial wireless networks, with a wireless transmitter at the sensor and a wireless receiver at the controller.

[0169] The memory stores the standard friction feature sequence corresponding to the standard valve, the anomaly judgment threshold, the valve model index, the detection condition information, and the computer program. The processor retrieves the corresponding standard friction feature sequence from the memory based on the valve model to be detected and the detection condition.

[0170] The test result output device is connected to the processor. The test result output device can be a display, status indicator, printer, or a communication interface connected to a host computer. The processor converts the assembly quality test results into display data, print data, or communication messages and then sends them to the test result output device.

[0171] When the processor executes the computer program, it first receives drive load data, valve displacement data, and time series data through the data acquisition interface; then it calculates the valve displacement change rate and drive load change rate, identifies nodes of motion state change, and divides the detection intervals into stages; subsequently, it establishes a load response curve, extracts friction feature parameters, forms a friction feature vector, and performs dimensionless normalization processing; then, it forms a segmented friction feature sequence according to the opening and closing stage sequence, and retrieves the corresponding standard friction feature sequence from memory; next, it determines the degree of feature deviation in each stage, compares the anomaly judgment threshold, and judges the friction anomaly state; finally, it determines the assembly defect type, anomaly level, and assembly quality state, generates assembly quality inspection results, and sends them to the inspection result output device.

[0172] A computer-readable storage medium storing a computer program, which, when executed by a processor, implements the valve assembly quality inspection method based on friction characteristics disclosed in the above embodiments. Computer devices include memory, processor, and computer programs stored in memory and capable of running on the processor.

[0173] The processor can be a central processing unit, microcontroller, digital signal processor, programmable logic controller, or other data processing device. The memory can include read-only memory, random access memory, solid-state memory, flash memory, or hard disk.

[0174] The computer program includes data acquisition instructions, stage division instructions, friction feature extraction instructions, feature sequence construction instructions, matching analysis instructions, and quality judgment instructions.

[0175] The data acquisition command is used to receive drive load data, valve displacement data, and time series data, and to perform time synchronization and preprocessing on the data.

[0176] The stage division instruction is used to calculate the valve displacement change rate and the drive load change rate, identify nodes of motion state change, and divide the stage detection interval.

[0177] The friction feature extraction command is used to establish a load response curve, determine at least one of the following: average load value, peak load value, load fluctuation amplitude, load change rate, and load change per unit displacement, and form a friction feature vector.

[0178] The feature sequence construction instruction is used to perform dimensionless normalization based on displacement stroke, stage duration and driving load reference value, and form a segmented friction feature sequence.

[0179] The matching analysis command is used to call the standard friction feature sequence, determine the degree of feature deviation in each opening and closing stage, and determine whether the corresponding stage is an abnormal friction state or a normal friction state.

[0180] The quality judgment instruction is used to determine the assembly defect type, abnormality level, and assembly quality status based on the correspondence between the abnormal stage, the opening and closing stage, and the assembly defect type, as well as the degree of feature deviation, and to generate assembly quality inspection results.

[0181] The computer equipment can be set up at the valve testing site and directly connected to the drive load sensor, displacement sensor and test result output device; or it can be connected to the valve assembly quality testing system through a communication network, and the on-site data acquisition equipment can send the drive load data, valve displacement data and time series data to the computer equipment for processing.

[0182] Compared with the prior art, the present invention has the following beneficial effects: (1) This invention obtains the driving load data, valve displacement data and time series data during the opening and closing process of the valve to be tested, and divides the opening and closing process into multiple stage detection intervals according to the valve displacement data, so that the entire valve opening and closing process is transformed from overall detection to stage detection, which can accurately locate the specific opening and closing stage where friction abnormality occurs, and improve the precision of assembly quality detection.

[0183] (2) The present invention extracts friction feature parameters for different detection intervals and constructs segmented friction feature sequences. By performing dimensionless normalization on the friction feature parameters of each stage, the influence of the differences in displacement stroke, duration and driving load magnitude of different stages on the detection results is reduced, the comparability between friction features of different opening and closing stages is improved, and the misjudgment and omission caused by using a unified judgment standard are reduced.

[0184] (3) This invention utilizes multiple valves of the same model with qualified assembly quality to establish a standard friction feature sequence under the same testing conditions, and performs a phased matching analysis between the segmented friction feature sequence corresponding to the valve to be tested and the standard friction feature sequence, so that the testing benchmark can correspond to the valve model and testing conditions, thereby improving the accuracy, stability and consistency of friction anomaly identification.

[0185] (4) Based on the friction abnormality state corresponding to each opening and closing stage, and combined with the correspondence between the opening and closing stage and the assembly defect type, the present invention can further determine the assembly defect type and abnormality level, so that the detection results can not only determine whether the valve under test has an assembly abnormality, but also output the abnormal stage, defect type, abnormality level and assembly quality status, providing a basis for assembly process optimization, defect tracing and product quality improvement.

[0186] (5) The present invention constitutes a valve assembly quality inspection system by driving load sensor, displacement sensor, controller and detection result output device, realizing automatic acquisition of drive load data and valve displacement data, automatic extraction of friction features, automatic matching of standard features and automatic judgment of assembly quality. It can meet the application needs of online detection and digital quality management of valve intelligent manufacturing production line, improve valve assembly quality inspection efficiency, reduce manual inspection error and improve product quality consistency.

[0187] (6) The computer equipment provided by the present invention can execute the above-mentioned valve assembly quality inspection method based on friction features, realize the software deployment of the inspection algorithm, and enable the present invention to be flexibly configured according to different valve models, inspection conditions and standard feature databases, and has good versatility, scalability and engineering application value.

[0188] In summary, this invention establishes a segmented friction feature analysis mechanism based on the division of opening and closing stages. By combining the changes in driving load during the valve opening and closing process with the valve's motion state, it achieves staged friction feature matching between the valve under test and the standard valve. This not only enables the location of friction anomalies but also allows for the identification of assembly defect types, the determination of anomaly levels, and the evaluation of assembly quality. This improves the accuracy, stability, and intelligence level of valve assembly quality inspection and is suitable for online inspection and digital quality control of assembly quality during valve manufacturing.

[0189] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A valve assembly quality inspection method based on friction characteristics, characterized in that, include: The driving load data, valve displacement data and corresponding time series data of the valve to be tested during the opening and closing process are obtained, and the opening and closing process is divided into multiple detection intervals based on the valve displacement data. For each of the aforementioned detection intervals, the friction characteristic parameters of the corresponding stage are determined based on the relationship between the driving load data and the valve displacement data, and a segmented friction characteristic sequence of the valve to be tested is constructed based on the friction characteristic parameters corresponding to different detection intervals. The segmented friction feature sequence is matched and analyzed with the standard friction feature sequence corresponding to the standard valve to determine the abnormal friction state of the valve under test in different opening and closing stages. The assembly quality status of the valve to be tested is determined based on the abnormal friction states during different opening and closing stages.

2. The valve assembly quality inspection method based on friction characteristics according to claim 1, characterized in that, Based on the valve displacement data, a single opening and closing process is divided into multiple detection intervals, including: The valve displacement change rate is determined based on the valve displacement data and the corresponding time series data, and the drive load change rate is determined based on the drive load data. Based on the coordinated relationship between the valve displacement change rate and the drive load change rate, identify the nodes of motion state change during the opening and closing process corresponding to static start-up, stable motion, sealing contact, and terminal pressing. According to the displacement position corresponding to each of the motion state change nodes, a single opening and closing process is divided into the starting disengagement stage, the stable motion stage, the sealing contact stage, and the terminal pressing stage. The start-up disengagement phase corresponds to the process by which the valve under test overcomes the initial static frictional resistance and begins to move; the stable movement phase corresponds to the frictional change process during the continuous movement of the valve under test; the sealing contact phase corresponds to the resistance change process after the sealing pair begins to contact; and the terminal pressing phase corresponds to the resistance change process after the sealing pair forms a pre-tightening force.

3. The valve assembly quality inspection method based on friction characteristics according to claim 1, characterized in that, The friction characteristic parameters for the corresponding stage are determined based on the relationship between the drive load data and the valve displacement data, including: Acquire drive load data, valve displacement data, and time series data within the corresponding detection interval; Based on the drive load data and the valve displacement data, a load response curve of the drive load as the valve displacement changes is established; At least one of the following parameters is extracted from the load response curve: average load value, peak load value, load fluctuation amplitude, load change rate, and load change per unit displacement, as the friction characteristic parameter for the corresponding stage. Friction feature vectors for the corresponding detection interval are formed based on the friction feature parameters of the corresponding stage.

4. The valve assembly quality inspection method based on friction characteristics according to claim 3, characterized in that, Based on the friction characteristic parameters corresponding to different detection intervals, a segmented friction characteristic sequence of the valve to be tested is constructed, including: Obtain the friction feature vectors corresponding to each of the detection intervals in each stage; Based on the displacement stroke, stage duration, and driving load reference value of the corresponding stage detection interval, the friction characteristic parameters in the friction characteristic vector of each stage are normalized without dimension. According to the sequence of each stage of the opening and closing process, the normalized friction feature vectors of each stage are combined sequentially to form a segmented friction feature sequence corresponding to one opening and closing process.

5. The valve assembly quality inspection method based on friction characteristics according to claim 4, characterized in that, Before performing matching analysis between the segmented friction feature sequence and the standard friction feature sequence corresponding to the standard valve, the following steps are also included: Acquire the drive load data, valve displacement data, and time series data of multiple qualified valves of the same model during the opening and closing process under the same testing conditions; For each valve that meets the assembly quality standards, the stage inspection interval is divided, friction feature parameters are extracted, and stage normalization is performed to obtain the segmented friction feature sequence corresponding to each valve that meets the assembly quality standards. Statistical analysis is performed on the multiple segmented friction feature sequences according to their corresponding opening and closing stages to determine the standard friction feature vector corresponding to each opening and closing stage. The standard friction feature vectors are combined according to the order of each stage of the opening and closing process to establish a standard friction feature sequence corresponding to the same type of valve, and the standard friction feature sequence is stored in the standard feature database.

6. The valve assembly quality inspection method based on friction characteristics according to claim 5, characterized in that, The segmented friction feature sequence is matched and analyzed with the standard friction feature sequence corresponding to the standard valve to determine the abnormal friction state of the valve under test in different opening and closing stages, including: The segmented friction feature sequence corresponding to the valve to be tested is matched with the standard friction feature sequence according to the opening and closing stages; Determine the degree of feature deviation between the friction feature vector of the valve to be tested and the corresponding standard friction feature vector in each opening and closing stage; The degree of deviation of the feature is compared with the anomaly determination threshold of the corresponding opening and closing stage; When the degree of deviation of the feature is greater than the abnormal judgment threshold of the corresponding opening and closing stage, it is determined that there is an abnormal friction state in the corresponding opening and closing stage; otherwise, it is determined that the corresponding opening and closing stage is a normal friction state. Based on the abnormal friction states corresponding to each opening and closing stage, the stage friction anomaly distribution results of the valve to be tested are formed.

7. The valve assembly quality inspection method based on friction characteristics according to claim 6, characterized in that, The assembly quality status of the valve under test is determined based on the abnormal friction states during different opening and closing stages, including: The abnormal stage is determined based on the abnormal friction state corresponding to each opening and closing stage. Based on the pre-established correspondence between the opening and closing stages and the assembly defect types, the assembly defect type corresponding to each abnormal stage is determined. Based on the assembly defect type and the degree of deviation of the corresponding opening and closing stage, the abnormality level of each assembly defect is determined. Based on the abnormality level and distribution of each assembly defect, the assembly quality status of the valve to be tested is determined, and the assembly quality test results are output.

8. A valve assembly quality inspection system based on friction characteristics, characterized in that, include: A load sensor is driven and installed at the drive mechanism of the valve to be tested; A displacement sensor is installed on the moving part of the valve to be detected; The controller includes a processor and a memory connected to the processor, the processor being electrically connected to the drive load sensor and the displacement sensor respectively; The detection result output device is connected to the processor; The memory stores a standard friction feature sequence and a computer program corresponding to a standard valve. When the processor executes the computer program, it implements the valve assembly quality detection method based on friction features as described in any one of claims 1-7, and sends the assembly quality detection result to the detection result output device.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the valve assembly quality inspection method based on friction characteristics as described in any one of claims 1-7.

10. A computer device, characterized in that, Includes a memory, a processor, and a computer program stored in the memory and capable of running on the processor; When the processor executes the computer program, it implements the valve assembly quality inspection method based on friction characteristics as described in any one of claims 1-7.

Images