Oil seal press-fitting abnormity real-time detection method based on pressure displacement curve analysis

Abstract

The invention provides an oil seal press-fitting abnormity real-time detection method based on pressure displacement curve analysis, and relates to the technical field of oil seal assembly control. Pressure displacement data acquisition is executed and a real-time pressure displacement curve is fitted when press-fitting equipment is started; based on displacement progress threshold triggering stage switching of the time sequence displacement data, an association stage characteristic curve is called from the oil seal press-fitting reference template; comparing the characteristic curve in the association stage with the real-time pressure displacement curve to solve a real-time curve deviation vector; and according to the real-time curve deviation vector mapping physical failure abnormal mode, self-adaptive adjustment of oil seal press-fitting process parameters is triggered. The technical problems that in the prior art, a continuous press-fitting process is adopted for oil seal assembly, rubber stress concentration is caused, hydraulic locking is formed due to lubricant retention, and the press-fitting depth is out of control are solved. Physical failure prevention in the oil seal assembly process is achieved, it is guaranteed that the press-fitting depth meets the technological specification tolerance in the whole process, oil seal assembly defects are restrained, and the sealing reliability of an engine is improved.

Description

Technical Field

[0001] This invention relates to the field of oil seal assembly control technology, and in particular to a real-time detection method for oil seal assembly anomalies based on pressure-displacement curve analysis. Background Technology

[0002] The existing oil seal assembly process uses a continuous compression process, which causes the molecular chains of the rubber material to not have time to rearrange during high-speed compression, resulting in huge stress concentration inside. This causes the sealing lip to twist, deform, or even flip, leading to initial seal failure and oil leakage.

[0003] At the same time, continuous pressing causes the lubricant to be squeezed rapidly between the sealing lip and the shaft, which cannot be discharged evenly and forms a local high-pressure oil wedge, resulting in a hydraulic locking effect. In severe cases, it can break the bond between the metal skeleton and the rubber or tear the rubber structure.

[0004] In addition, high-speed impacts cause uncontrolled pressing depth, and inertia causes the oil seal to be over-pressed or under-pressed, which seriously affects the quality of engine assembly. These chain-reaction physical defects ultimately lead to premature aging and failure of the oil seal and loss of sealing function.

[0005] In summary, the existing technology using continuous press-fitting process for oil sealing has technical problems such as rubber stress concentration, lubricant retention leading to hydraulic lock-up, and uncontrolled press-fitting depth. Summary of the Invention

[0006] To address the aforementioned deficiencies or improvement needs of existing technologies, this invention provides a real-time detection method for oil seal press-fitting anomalies based on pressure-displacement curve analysis. This method solves the technical problems of existing technologies using continuous press-fitting processes for oil seal assembly, which can lead to rubber stress concentration, lubricant retention causing hydraulic lock-up, and uncontrolled press-fitting depth.

[0007] To achieve the above objectives, this invention provides a real-time detection method for oil seal press-fitting anomalies based on pressure-displacement curve analysis, the method comprising:

[0008] By collecting standard oil seal pressing operation data, an oil seal pressing benchmark template is constructed. This template includes standard characteristic curves for the rough pressing stage, the pause stage, and the fine pressing stage. When the pressing equipment starts, pressure and displacement sensors are triggered to synchronously collect data, outputting time-series pressure and displacement data. A real-time pressure-displacement curve is fitted based on the time-series pressure and displacement data. A stage switching is triggered based on the displacement progress threshold of the time-series displacement data, and the associated stage characteristic curve is retrieved from the oil seal pressing benchmark template. The associated stage characteristic curve and the real-time pressure-displacement curve are compared, and the real-time curve deviation vector is solved using a dynamic time warping algorithm. A physical failure anomaly mode is mapped based on the real-time curve deviation vector. An adaptive adjustment of the oil seal pressing process parameters is triggered according to the physical failure anomaly mode.

[0009] In one implementation, by collecting standard oil seal press-fitting condition data, an oil seal press-fitting benchmark template is constructed, and the following processing is also performed:

[0010] Under the constraint of standard oil seal press-fitting conditions, data acquisition of the oil seal press-fitting process is performed to obtain multi-cycle pressure-displacement time-series data. By performing segmented feature extraction on the multi-cycle pressure-displacement time-series data, an oil seal press-fitting benchmark template is constructed. The standard feature curve of the rough pressing stage is identified by the peak pressure threshold range at the end of the rough pressing stage, the standard feature curve of the pause stage is identified by the pressure attenuation slope threshold range, and the standard feature curve of the fine pressing stage is identified by the displacement increment threshold range.

[0011] In one implementation, based on the physical failure anomaly mode mapped by the real-time curve deviation vector, the following processing is also performed:

[0012] The system interactively obtains multiple sample curve deviation vectors, multiple sample pressure change rate intervals, and multiple sample displacement change rate intervals for various predefined abnormal failure modes. Based on the physical mechanism of oil seal failure, it associates and binds these predefined abnormal failure modes, sample curve deviation vectors, sample pressure change rate intervals, and sample displacement change rate intervals to complete the construction of a physical failure mode identification model. It traverses the real-time pressure and displacement curves to solve for real-time pressure change rate features and real-time displacement change rate features. It loads the real-time curve deviation vectors, real-time pressure change rate features, and real-time displacement change rate features into the physical failure mode identification model to calculate pattern matching confidence and outputs the physical failure abnormal mode, wherein the physical failure abnormal mode is marked with an abnormal confidence level.

[0013] In one implementation, based on the displacement progress threshold triggered by the time-series displacement data, the associated stage characteristic curve is retrieved from the oil seal press-fitting reference template, and the following processing is also performed:

[0014] A preset oil seal press-fit displacement threshold is established, which includes a rough press-fit endpoint threshold, a pause confirmation threshold, and a fine press-fit start threshold. Feature extraction is performed on the time-series displacement data to obtain the press-fit stroke percentage and displacement stabilization time window. The press-fit stroke percentage and displacement stabilization time window are used to iterate through the oil seal press-fit displacement threshold to match and locate the real-time execution stage. Based on the real-time execution stage, the associated stage feature curve is retrieved from the oil seal press-fit reference template.

[0015] In one implementation, by comparing the associated stage characteristic curve and the real-time pressure-displacement curve, and solving the real-time curve deviation vector using a dynamic time warping algorithm, the following processing is also performed:

[0016] The correlation stage feature curve and the real-time pressure-displacement curve are normalized based on the displacement percentage; the time-series pressure data points of the correlation stage feature curve and the real-time pressure-displacement curve are matched based on the dynamic time warping algorithm to generate the optimal alignment path; the pressure value difference is calculated point by point along the optimal alignment path to generate the initial curve deviation vector; the initial curve deviation vector is weighted by physical feature enhancement according to the real-time execution stage to output the real-time curve deviation vector.

[0017] In one implementation, the following processing is also performed:

[0018] If the real-time execution stage is the coarse pressure stage, then after the real-time pressure-displacement curve passes through the pressure value range of the coarse pressure endpoint pressure peak threshold interval, a deviation comparison is performed between the standard characteristic curve of the coarse pressure stage and the real-time pressure-displacement curve. If the real-time execution stage is the pause stage, then after the real-time pressure-displacement curve passes through the slope value range of the pressure attenuation slope threshold interval, a deviation comparison is performed between the standard characteristic curve of the pause stage and the real-time pressure-displacement curve. If the real-time execution stage is the fine pressure stage, then after the real-time pressure-displacement curve passes through the displacement value range of the displacement increment threshold interval, a deviation comparison is performed between the standard characteristic curve of the fine pressure stage and the real-time pressure-displacement curve.

[0019] In one implementation, the adaptive adjustment of the oil seal press-fitting process parameters is triggered based on the physical failure anomaly mode, and the following processing is also performed:

[0020] The physical failure anomaly mode is analyzed to obtain the anomaly type identifier and the anomaly confidence level; the anomaly type identifier is used as a search condition to call the real-time adjustment strategy in the predefined strategy mapping library, wherein the strategy mapping library stores multiple sample adjustment strategies for multiple sample anomaly types; after correcting the oil seal press-fitting process parameters using the real-time adjustment strategy, the oil seal press-fitting closed-loop process compensation is executed according to the time-series pressure increment and time-series displacement increment.

[0021] In one implementation, if the anomaly confidence level is lower than a preset confidence threshold, the pressing process is paused and a manual re-inspection instruction is triggered.

[0022] One or more technical solutions provided in this invention have at least the following technical effects or advantages:

[0023] The method provided in this invention constructs an oil seal press-fitting benchmark template by collecting standard oil seal press-fitting data, including standard characteristic curves for the rough pressing stage, pause stage, and fine pressing stage. When the press-fitting equipment starts, pressure and displacement sensors are simultaneously triggered to collect data and fit real-time pressure-displacement curves. Based on a displacement progress threshold, stage switching is triggered, and the associated stage characteristic curves in the benchmark template are retrieved. A dynamic time warping algorithm is used to compare the real-time curves with the associated characteristic curves to generate a real-time curve deviation vector. This vector is used to map physical failure anomaly modes and trigger adaptive adjustment of process parameters. Ultimately, this achieves physical failure prevention of rubber stress concentration and lubricant retention during the oil seal assembly process, ensuring that the press-fitting depth meets process specification tolerances throughout the process, completely suppressing sealing lip distortion and skeleton debonding defects, and fundamentally improving the long-term sealing reliability of the engine. This method achieves the technical effects of physical failure prevention in the oil seal assembly process, ensuring that the press-fitting depth meets process specification tolerances throughout the process, suppressing oil seal assembly defects, and improving engine sealing reliability. Attached Figure Description

[0024] 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.

[0025] Figure 1 A schematic diagram of the real-time detection method for oil seal press-fitting anomalies based on pressure-displacement curve analysis provided by the present invention is shown.

[0026] Figure 2 The diagram illustrates the process of solving the real-time curve deviation vector in the real-time detection method for oil seal press-fitting anomalies based on pressure-displacement curve analysis provided by the present invention. Detailed Implementation

[0027] This invention provides a real-time detection method for oil seal press-fitting anomalies based on pressure-displacement curve analysis, which is used to solve the technical problems of existing oil seal assembly processes using continuous press-fitting, such as rubber stress concentration, lubricant retention leading to hydraulic lock-up, and uncontrolled press-fitting depth.

[0028] 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. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0029] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0030] Unless otherwise expressly stated, throughout the specification and claims, the term "comprising" or its variations such as "including" or "comprises" shall be understood to include the stated elements or components without excluding other elements or other components.

[0031] Example 1: A flowchart of the real-time detection method for oil seal press-fitting anomalies based on pressure-displacement curve analysis provided in this embodiment of the invention. (See attached diagram) Figure 1 The method includes:

[0032] A100: By collecting standard oil seal press-fitting working condition data, an oil seal press-fitting benchmark template is constructed, wherein the oil seal press-fitting benchmark template includes standard characteristic curves for the rough pressing stage, standard characteristic curves for the pause stage, and standard characteristic curves for the fine pressing stage.

[0033] In one embodiment, by collecting standard oil seal press-fitting condition data, an oil seal press-fitting benchmark template is constructed. The method step A100 provided by this invention further includes:

[0034] A110: Using the standard oil seal press-fitting conditions as constraints, perform oil seal press-fitting process data acquisition to obtain multi-cycle pressure displacement time series data.

[0035] A120: By performing segmented feature extraction on the multi-cycle pressure-displacement time series data, an oil seal press-fitting reference template is constructed. The standard feature curve of the rough pressing stage is identified by the peak pressure threshold range at the end of the rough pressing stage, the standard feature curve of the pause stage is identified by the pressure attenuation slope threshold range, and the standard feature curve of the fine pressing stage is identified by the displacement increment threshold range.

[0036] Specifically, the standard oil seal press-fitting condition refers to a defect-free assembly process that conforms to the process specifications. In this embodiment, the standard oil seal press-fitting condition is used as a constraint. During the oil seal press-fitting process, the physical parameters of the entire assembly process are collected synchronously by pressure and displacement sensors. The multi-cycle pressure-displacement time series data, which characterizes the pressure-displacement changes over time in multiple qualified assemblies, is obtained, providing a statistical basis for the oil seal press-fitting benchmark template.

[0037] By dividing the multi-cycle pressure displacement time series data into three stages—coarse pressing, pause, and fine pressing—an oil seal press fitting reference template is constructed. The oil seal press fitting reference template specifically includes the standard characteristic curve of the coarse pressing stage, which reflects the initial compression pressure change characteristics; the standard characteristic curve of the pause stage, which reflects the pressure decay characteristics during the stress relaxation period; and the standard characteristic curve of the fine pressing stage, which characterizes the displacement control characteristics of the precision press fitting.

[0038] The standard characteristic curve for the roughing stage is marked by the peak pressure threshold range at the end of the roughing stage, which is used to identify the critical pressure range at which rubber compression is completed. The standard characteristic curve for the pause stage is marked by the pressure decay slope threshold range, which is used to identify the reasonable decay rate of lubricant discharge, and is related to the lubricant viscosity and discharge time. The standard characteristic curve for the fine pressing stage is marked by the displacement increment threshold range, which is used to identify the allowable stroke error in the fine pressing stage to ensure the accuracy of the pressing depth. The threshold ranges of the above three stages are the quantitative carriers of the process physical mechanism.

[0039] This implementation constructs an oil seal press-fitting benchmark template as a process quality benchmark, which is used as a basis for curve comparison in real-time detection.

[0040] A200: When the press-fitting equipment starts, it triggers the pressure sensor and displacement sensor to synchronously perform data acquisition and output time-series pressure data and time-series displacement data.

[0041] A300: Fit a real-time pressure-displacement curve based on the time-series pressure data and time-series displacement data.

[0042] Specifically, in this embodiment, when the pressing equipment is started, the pressure sensor and displacement sensor are triggered to synchronously perform data acquisition, and output the time-series pressure data reflecting the dynamic changes in assembly resistance, and the time-series displacement data characterizing the pressing depth advancement process.

[0043] The real-time pressure-displacement curve obtained by fitting the time-series pressure data and time-series displacement data is a two-dimensional plane curve. The curve uses displacement data as the horizontal axis and pressure data as the vertical axis to form a displacement-pressure relationship curve. This curve intuitively presents the evolution trajectory of the mechanical state as the pressing depth changes during the assembly process.

[0044] For example, the real-time pressure-displacement curve provides a visual basis for subsequent stage switching and anomaly detection, as the pressure suddenly increases when the lubricant is retained or decreases when the rubber stress is released.

[0045] A400: Based on the displacement progress threshold of the time-series displacement data, the stage switching is triggered, and the associated stage characteristic curve is retrieved from the oil seal press-fit reference template.

[0046] In one embodiment, a stage switching is triggered based on a displacement progress threshold of the time-series displacement data, and a related stage characteristic curve is retrieved from the oil seal press-fitting reference template. Step A400 of the method provided by this invention further includes:

[0047] A410: Preset oil seal press-fit displacement threshold, wherein the oil seal press-fit displacement threshold includes a rough press end point threshold, a pause confirmation threshold, and a fine press start point threshold.

[0048] A420: Perform feature extraction on the time-series displacement data to obtain the percentage of pressing stroke and displacement stabilization time window.

[0049] A430: The oil seal pressing displacement threshold is traversed using the pressing stroke percentage and displacement stabilization time window to match and locate the real-time execution stage.

[0050] A440: Based on the real-time execution stage, retrieve the associated stage feature curve from the oil seal press-fit reference template.

[0051] Specifically, this embodiment predefines three-stage conversion preset displacement triggering conditions as the oil seal press-fit displacement threshold, which includes the rough press end threshold, the pause confirmation threshold, and the fine press start threshold.

[0052] The roughing end threshold corresponds to the point where the rubber is initially compressed and the displacement reaches 90%-95% of the total stroke. The pause confirmation threshold is used to ensure the stability of the stress relaxation period, with the displacement fluctuation ≤0.02mm and lasting ≥1.5 seconds. The fine pressing start threshold is used to reference and trigger the precision pressing, with the displacement continuing to advance from 95% of the total stroke.

[0053] In this embodiment, a predefined coarse compression endpoint threshold is used to lock the rubber compression completion position, a pause confirmation threshold is used to verify the stress relaxation period through displacement stability, and a fine compression starting threshold is used to initiate precision control.

[0054] Feature extraction is performed on the time-series displacement data to obtain the percentage of pressing stroke used to determine whether it is close to the 90% rough pressing endpoint, and the displacement stabilization time window used to verify the effectiveness of the pause phase.

[0055] The current stage is located by matching the coarse pressure end point / fine pressure start point threshold with the real-time stroke percentage and matching the pause confirmation threshold with the displacement stabilization time window.

[0056] For example, if the stroke percentage is ≥90% and the stable time window is ≥1.5 seconds, then the real-time execution stage is a pause stage; if the stroke percentage is ≥95% and the displacement resumes to grow, then the real-time execution stage is a fine-pressing stage.

[0057] Based on the real-time execution phase, the characteristic curve of the associated phase is retrieved from the oil seal pressing reference template, which achieves the technical effect of providing a physical quantitative reference for subsequent process anomaly detection.

[0058] A500: By comparing the characteristic curve of the associated stage with the real-time pressure-displacement curve, the real-time curve deviation vector is solved by the dynamic time warping algorithm.

[0059] In one implementation, see Figure 2 By comparing the associated stage characteristic curve and the real-time pressure-displacement curve, and solving the real-time curve deviation vector using a dynamic time warping algorithm, the method step A500 provided by this invention further includes:

[0060] A510: Normalize the characteristic curves of the associated stage and the real-time pressure-displacement curves based on the displacement percentage.

[0061] A520: Based on the dynamic time warping algorithm, the time-series pressure data points of the associated stage feature curve and the real-time pressure-displacement curve are matched to generate the optimal alignment path.

[0062] A530: Traverse the associated stage feature curve and real-time pressure-displacement curve along the optimal alignment path to calculate the pressure value difference point by point and generate an initial curve deviation vector.

[0063] A540: Perform physical feature enhancement weighting on the initial curve deviation vector according to the real-time execution phase, and output the real-time curve deviation vector.

[0064] In one embodiment, the method provided by the present invention further includes:

[0065] A5001: If the real-time execution stage is the coarse pressure stage, then after the real-time pressure displacement curve passes through the pressure value range of the coarse pressure endpoint pressure peak threshold interval, the deviation comparison between the standard characteristic curve of the coarse pressure stage and the real-time pressure displacement curve is performed.

[0066] A5002: If the real-time execution phase is a pause phase, then after the real-time pressure displacement curve passes through the slope value range of the pressure attenuation slope threshold interval, the deviation comparison between the standard characteristic curve of the pause phase and the real-time pressure displacement curve is performed.

[0067] A5003: If the real-time execution stage is the fine-pressure stage, then after the real-time pressure-displacement curve passes the displacement value range of the displacement increment threshold interval, the deviation comparison between the standard characteristic curve of the fine-pressure stage and the real-time pressure-displacement curve is performed.

[0068] Before comparing the correlation stage feature curve and the real-time pressure-displacement curve, the real-time pressure-displacement curve is first subjected to physical threshold rapid interception verification.

[0069] Specifically, if the real-time execution stage is the coarse pressure stage, it is necessary to verify whether the coarse pressure endpoint peak value of the real-time pressure displacement curve is within the safe range of the coarse pressure endpoint peak value threshold range.

[0070] This verification determines the pressure value range within the displacement stroke of 90% to 95%. If the pressure value exceeds the upper limit of the peak pressure threshold range at the end of the coarse compression, it indicates that stress concentration is generated during the compression process of the rubber molecular chains, directly triggering the abnormal interception mechanism. If the pressure value is within the range of the peak pressure threshold range at the end of the coarse compression, the curve deviation comparison is performed to further monitor the stress distribution.

[0071] If the real-time execution phase is a pause phase, it is necessary to verify whether the pressure decay rate of the real-time pressure displacement curve meets reasonable requirements.

[0072] This verification calculates the slope of pressure change during the displacement stabilization period. If the decay rate does not reach the lower limit of the pressure decay slope threshold range, it indicates a risk of lubricant retention leading to hydraulic lock-up, and the curve comparison is skipped, directly triggering an anomaly. If the decay rate meets the requirements of the pressure decay slope threshold range, it indicates that the lubricant discharge dynamics are normal, and the uniformity of pressure decay is then monitored through curve comparison. This step is closely related to the rheological properties of the lubricant, ensuring that the lubricant is fully and uniformly discharged during the two-second pause period.

[0073] If the real-time execution stage is the precision pressing stage, it is necessary to verify whether the real-time displacement propulsion is within the accuracy tolerance range.

[0074] This verification measures the displacement increment within the 95% to 100% displacement travel range. If the increment exceeds the upper limit of the displacement increment threshold range, it indicates an overpressure risk in the precision press-fitting process, and the assembly process is immediately terminated. If the increment is insufficient, it is marked as underpressure, but pressure stability still needs to be verified by curve comparison. This step directly serves the oil seal press-fitting depth control target, preventing skeleton detachment or poor sealing lip fit.

[0075] After verifying the status at the corresponding stage and excluding the failure of the phased locking oil seal assembly, the real-time curve deviation vector is solved by the dynamic time warping algorithm.

[0076] Specifically, in this embodiment, the displacement coordinates of the associated stage characteristic curve and the real-time pressure-displacement curve are uniformly converted into percentage values ​​relative to the total stroke to eliminate the influence of oil seal size differences on the curve shape, ensure the comparability of pressure-displacement curves of different specifications of products, and provide standardized input for the dynamic time warping algorithm.

[0077] The dynamic time warping algorithm is a nonlinear timing matching method that aligns the pressure data points of two curves through local scaling, eliminating timing deviations caused by equipment response delays or speed fluctuations. The optimal alignment path is the best correspondence sequence of pressure points between the two curves generated by the dynamic time warping algorithm.

[0078] This embodiment employs the dynamic time warping algorithm, using displacement percentage as a benchmark, and under the premise of allowing local compression / stretching, to find the minimum cumulative distance mapping path between the characteristic curve of the associated stage and the pressure point of the real-time pressure-displacement curve as the optimal alignment path.

[0079] The optimal alignment path precisely aligns with the stress relaxation inflection point or pressure peak point, ensuring that subsequent difference calculations are based on the same physical state.

[0080] For each matching point along the optimal alignment path, the pressure deviation value between the associated stage characteristic curve and the real-time pressure-displacement curve at the same displacement percentage is calculated. The resulting initial curve deviation vector is a sequence composed of the pressure deviation values ​​of all matching points in order of displacement percentage.

[0081] The initial curve deviation vector quantifies the microscopic differences between the real-time assembly process and the benchmark. Specifically, a positive deviation indicates excessive assembly resistance, such as lubricant retention; a negative deviation indicates insufficient resistance, such as lack of lubrication.

[0082] For example, if the real-time pressure is 0.3 MPa higher than the standard value at the 90% displacement point during the roughing stage, it indicates a risk of rubber stress concentration.

[0083] The initial curve deviation vector is weighted by physical feature enhancement during the real-time execution phase, and the real-time curve deviation vector is output.

[0084] The specific weighting strategies include: increasing the weight of the displacement range of 90%-95% during the coarse pressing stage, increasing the weight of the pressure attenuation window during the pause stage, and increasing the weight of the displacement range of 95%-100% during the fine pressing stage.

[0085] The weighted deviation vector amplifies the key abnormal signal and weakens the noise in the non-sensitive area, so that the output result closely matches the failure physical mechanism of the oil package assembly.

[0086] A600: Based on the real-time curve deviation vector mapping physical failure anomaly mode.

[0087] In one embodiment, based on the real-time curve deviation vector mapping physical failure anomaly mode, the method step A600 provided by the present invention further includes:

[0088] A610: Interactively obtain multiple sample curve deviation vectors, multiple sample pressure change rate intervals, and multiple sample displacement change rate intervals for various predefined abnormal failure modes.

[0089] A620: Based on the physical mechanism of oil seal failure, the model is constructed by associating and binding multiple predefined abnormal failure modes, multiple sample curve deviation vectors, multiple sample pressure change rate intervals, and multiple sample displacement change rate intervals.

[0090] A630: Traverse the real-time pressure-displacement curves to solve for the real-time pressure change rate characteristics and the real-time displacement change rate characteristics.

[0091] A640: Load the real-time curve deviation vector, real-time pressure change rate feature, and real-time displacement change rate feature into the physical failure mode recognition model to calculate the pattern matching confidence level, and output the physical failure anomaly mode, wherein the physical failure anomaly mode has an anomaly confidence level identifier.

[0092] This embodiment identifies physical anomalies such as rubber stress concentration and lubricant retention by comparing real-time pressure deviation characteristics with a predefined failure mode feature library, providing a basis for adaptive process adjustments. Its technical essence is to correlate the microscopic deformation of the pressure-displacement curve with the failure physical mechanism of the oil-sealed assembly, achieving a closed loop from sensor data to process decisions.

[0093] Specifically, in this embodiment, a physical failure mode identification model covering all failure scenarios is pre-constructed. The physical failure mode identification model stores multiple sets of predefined abnormal failure modes, sample curve deviation vectors, sample pressure change rate intervals, and sample displacement change rate intervals that are associated and bound based on the physical mechanism of oil seal failure.

[0094] Among them, the predefined abnormal failure mode refers to the fault classification predefined based on the failure mechanism of oil seal assembly, the sample curve deviation vector is the pressure deviation feature sequence of known abnormal types, the sample pressure change rate interval represents the reasonable range of pressure change rate when the abnormality occurs, such as the attenuation rate when lubricant is retained > -0.1MPa / s, and the sample displacement change rate interval represents the abnormal displacement propagation rate, such as the displacement increment > 0.05mm / 0.1s when overpressure occurs.

[0095] The physical failure mode identification model is a classifier that includes mechanistic constraints. Its output is strictly limited by the mechanical and fluid dynamic principles of oil seal materials to avoid misjudgments that do not conform to physical laws.

[0096] The real-time pressure change rate characteristic and the real-time displacement change rate characteristic are obtained by traversing the real-time pressure-displacement curve. The real-time pressure change rate characteristic is the first derivative of the real-time pressure-displacement curve, which quantifies the rate of pressure change over time, with the unit being MPa / s, and directly reflects the lubricant discharge efficiency. The real-time displacement change rate characteristic is the rate of displacement change over time, with the unit being mm / s, which characterizes the stability of the press-fitting process. For example, a change rate > 0.5 mm / s during the fine pressing stage indicates the risk of overshoot.

[0097] A640: Load the real-time curve deviation vector, real-time pressure change rate feature, and real-time displacement change rate feature into the physical failure mode recognition model to calculate the pattern matching confidence level, and output the physical failure anomaly mode, wherein the physical failure anomaly mode has an anomaly confidence level identifier.

[0098] Pattern matching confidence calculation involves inputting the real-time curve deviation vector, real-time pressure change rate, and real-time displacement change rate into the physical failure mode recognition model. By comparing real-time features with a predefined sample abnormal feature library, a comprehensive diagnosis is performed, and the specific abnormal type and its probability estimate are output.

[0099] Specifically, the real-time curve deviation vector, real-time pressure change rate, and real-time displacement change rate are traversed and calculated along with multiple sets of bound data in the physical failure mode identification model. Each set of bound data includes the sample curve deviation vector, pressure change rate range, and displacement change rate range, and is strictly bound to the physical mechanism of an abnormal failure mode.

[0100] The similarity between real-time features and each set of bound data is quantified using Euclidean distance to obtain the state similarity value corresponding to each predefined abnormal failure mode.

[0101] After sorting all state similarities in descending order, the predefined abnormal failure mode corresponding to the maximum value is extracted as the physical failure abnormal mode, and the maximum similarity value is output as the abnormal confidence index.

[0102] This embodiment achieves the technical effect of accurate failure mode identification, providing a physically explainable decision-making basis for subsequent adaptive adjustment of process parameters.

[0103] A700: Triggers adaptive adjustment of oil seal press-fit process parameters based on the physical failure anomaly mode.

[0104] In one embodiment, the method step A700 of the present invention further includes: Based on the physical failure anomaly mode triggering adaptive adjustment of the oil seal press-fitting process parameters, the method step A700 of the present invention further includes:

[0105] A710: Parse the physical failure anomaly mode to obtain the anomaly type identifier and the anomaly confidence level.

[0106] A720: Using the anomaly type identifier as a search condition, a real-time adjustment strategy is invoked from a predefined strategy mapping library, wherein the strategy mapping library stores multiple sample adjustment strategies for multiple sample anomaly types.

[0107] A730: After correcting the oil seal press-fitting process parameters using the real-time adjustment strategy, the oil seal press-fitting closed-loop process compensation is performed based on the time-series pressure increment and time-series displacement increment.

[0108] In one implementation, if the anomaly confidence level is lower than a preset confidence threshold, the pressing process is paused and a manual re-inspection instruction is triggered.

[0109] Specifically, this embodiment obtains the anomaly type identifier and the anomaly confidence level by parsing the physical failure anomaly mode. The anomaly type identifier marks the specific failure category, such as stress concentration, lubricant lock-up, and press-fitting deviation, while the anomaly confidence level quantifies the reliability of the diagnosis.

[0110] The preset confidence threshold is the critical value for determining the reliability of the diagnosis. If the abnormal confidence level is lower than this value, it indicates that sensor noise or occasional interference may lead to misjudgment. At this time, the pressing process is paused and a manual re-inspection instruction is triggered to prompt manual inspection of the lip lubricant distribution and avoid erroneous adjustments that could interfere with process stability.

[0111] Conversely, if the anomaly confidence level is higher than the preset confidence threshold, the anomaly type identifier is used as a retrieval condition, and a sample adjustment strategy corresponding to a sample anomaly type that is consistent with the anomaly type identifier is called from the predefined strategy mapping library as the real-time adjustment strategy.

[0112] It should be understood that the strategy mapping library is an index database, which stores multiple sample adjustment strategies for multiple sample anomaly types.

[0113] A730: After correcting the oil seal press-fitting process parameters using the real-time adjustment strategy, the oil seal press-fitting closed-loop process compensation is performed based on the time-series pressure increment and time-series displacement increment.

[0114] After adopting a real-time adjustment strategy to correct the oil seal press-fitting process parameters, the stress release status is fed back in real time by continuously monitoring the time-series pressure increment, and the press-fitting depth accuracy is verified in real time by tracking the time-series displacement increment, thus forming a closed-loop process compensation mechanism.

[0115] The process dynamically verifies the effectiveness of the adjustment strategy. If the pressure increment recovers to the standard attenuation rate range, the stress relaxation is confirmed to be up to standard. If the displacement increment meets the preset accuracy tolerance, the pressing depth is judged to be qualified.

[0116] When the pressure increment meets the standard but the displacement increment is abnormal, the secondary fine pressure thrust correction is automatically triggered; if the displacement increment is qualified but the pressure increment has not recovered, the pause time is extended again until the stress release and pressing depth requirements are met simultaneously, so as to realize the adaptive process optimization of the oil seal assembly process.

[0117] This embodiment achieves physical failure prevention in the oil seal assembly process through phased dynamic monitoring and closed-loop control, ensuring that the pressing depth meets the process specification tolerance throughout the process, suppressing oil seal assembly defects and improving the reliability of engine sealing.

[0118] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A real-time detection method for oil seal press-fitting anomalies based on pressure-displacement curve analysis, characterized in that, include: By collecting standard oil seal press-fitting working condition data, an oil seal press-fitting benchmark template is constructed, wherein the oil seal press-fitting benchmark template includes standard characteristic curves for the rough pressing stage, standard characteristic curves for the pause stage, and standard characteristic curves for the fine pressing stage. When the press-fitting equipment starts, the pressure sensor and displacement sensor are triggered to synchronously perform data acquisition and output time-series pressure data and time-series displacement data. A real-time pressure-displacement curve is fitted based on the time-series pressure data and time-series displacement data; Based on the displacement progress threshold triggered by the time-series displacement data, the associated stage feature curve is retrieved from the oil seal press-fitting reference template. By comparing the characteristic curves of the associated stage with the real-time pressure-displacement curves, the real-time curve deviation vector is solved using a dynamic time warping algorithm. Based on the real-time curve deviation vector mapping, physical failure anomaly modes are observed; The adaptive adjustment of oil seal press-fitting process parameters is triggered based on the physical failure anomaly mode.

2. The real-time detection method for oil seal press-fitting anomalies based on pressure-displacement curve analysis as described in claim 1, characterized in that, By collecting standard oil seal press-fitting condition data, a benchmark template for oil seal press-fitting is constructed, including: Under the constraint of standard oil seal press-fitting conditions, data acquisition of the oil seal press-fitting process was performed to obtain multi-cycle pressure displacement time series data; By segmenting and extracting features from the multi-cycle pressure-displacement time series data, an oil seal press-fitting reference template is constructed. The standard feature curve of the rough pressing stage is identified by the peak pressure threshold range at the end of the rough pressing stage, the standard feature curve of the pause stage is identified by the pressure attenuation slope threshold range, and the standard feature curve of the fine pressing stage is identified by the displacement increment threshold range.

3. The real-time detection method for oil seal press-fitting anomalies based on pressure-displacement curve analysis as described in claim 1, characterized in that, Based on the real-time curve deviation vector mapping physical failure anomaly mode, including: Interactively obtain multiple sample curve deviation vectors, multiple sample pressure change rate intervals, and multiple sample displacement change rate intervals for various predefined abnormal failure modes; Based on the physical mechanism of oil seal failure, the physical failure mode identification model is constructed by associating and binding multiple predefined abnormal failure modes, multiple sample curve deviation vectors, multiple sample pressure change rate intervals, and multiple sample displacement change rate intervals. The real-time pressure change rate characteristics and real-time displacement change rate characteristics are solved by traversing the real-time pressure-displacement curves. The real-time curve deviation vector, real-time pressure change rate feature, and real-time displacement change rate feature are loaded into the physical failure mode recognition model to calculate the pattern matching confidence level, and the physical failure anomaly mode is output, wherein the physical failure anomaly mode is marked with an anomaly confidence level.

4. The real-time detection method for oil seal press-fitting anomalies based on pressure-displacement curve analysis as described in claim 2, characterized in that, Based on the displacement progress threshold triggered by the time-series displacement data, the stage switching is performed, and the associated stage characteristic curve is retrieved from the oil seal press-fitting reference template, including: A preset oil seal press-fit displacement threshold is provided, wherein the oil seal press-fit displacement threshold includes a rough press end point threshold, a pause confirmation threshold, and a fine press start point threshold. Feature extraction is performed on the time-series displacement data to obtain the percentage of pressing stroke and displacement stabilization time window; The oil seal press-fit displacement threshold is traversed using the press-fit stroke percentage and displacement stabilization time window to match and locate the real-time execution stage. Based on the real-time execution phase, the characteristic curve of the associated phase is retrieved from the oil seal press-fit reference template.

5. The real-time detection method for oil seal press-fitting anomalies based on pressure-displacement curve analysis as described in claim 4, characterized in that, By comparing the associated stage characteristic curve and the real-time pressure-displacement curve, the real-time curve deviation vector is solved using a dynamic time warping algorithm, including: The correlation stage characteristic curve and real-time pressure-displacement curve are normalized based on the displacement percentage. The optimal alignment path is generated by matching the time-series pressure data points of the associated stage feature curve and the real-time pressure-displacement curve based on the dynamic time warping algorithm. The pressure value difference is calculated point by point along the optimal alignment path by traversing the associated stage feature curve and the real-time pressure-displacement curve to generate an initial curve deviation vector. The initial curve deviation vector is weighted by physical feature enhancement during the real-time execution phase, and the real-time curve deviation vector is output.

6. The real-time detection method for oil seal press-fitting anomalies based on pressure-displacement curve analysis as described in claim 5, characterized in that, Also includes: If the real-time execution stage is the coarse pressure stage, then after the real-time pressure displacement curve passes through the pressure value range of the pressure peak threshold interval of the coarse pressure endpoint, the deviation comparison between the standard characteristic curve of the coarse pressure stage and the real-time pressure displacement curve is performed. If the real-time execution phase is a pause phase, then after the real-time pressure displacement curve passes through the slope value range of the pressure attenuation slope threshold interval, the deviation comparison between the standard characteristic curve of the pause phase and the real-time pressure displacement curve is performed. If the real-time execution stage is the fine-pressure stage, then after the real-time pressure-displacement curve passes the displacement value range of the displacement increment threshold interval, the deviation comparison between the standard characteristic curve of the fine-pressure stage and the real-time pressure-displacement curve is performed.

7. The real-time detection method for oil seal press-fitting anomalies based on pressure-displacement curve analysis as described in claim 3, characterized in that, The adaptive adjustment of oil seal press-fitting process parameters is triggered based on the aforementioned physical failure anomaly mode, including: Analyze the physical failure anomaly mode to obtain the anomaly type identifier and the anomaly confidence level; The anomaly type identifier is used as a search condition to call a real-time adjustment strategy in a predefined strategy mapping library, wherein the strategy mapping library stores multiple sample adjustment strategies for multiple sample anomaly types. After correcting the oil seal press-fitting process parameters using the aforementioned real-time adjustment strategy, closed-loop process compensation for oil seal press-fitting is performed based on the time-series pressure increment and time-series displacement increment.

8. The real-time detection method for oil seal press-fitting anomalies based on pressure-displacement curve analysis as described in claim 7, characterized in that, If the anomaly confidence level is lower than the preset confidence threshold, the pressing process is paused and a manual re-inspection instruction is triggered.

Images