A method and apparatus for processing data of a graph estimation test of a seal group reliability

Abstract

The application provides a kind of seal group reliability test graph estimation data processing method and device, belongs to seal field.The method provided by the application comprises: regarding each group of seals as a sample, based on the distribution characteristics of seal failure life data, the distribution parameters are calculated by graph estimation, and the basic fitting curve is drawn;Combined with the number of components in the group and the test stop rule of seal group test, the offset processing is carried out on the basic fitting curve, and the reliability estimation graph corresponding to the group test is drawn;Based on the reliability estimation graph, the reliability analysis parameters of seal life test are calculated, and the reliability of the seal is evaluated based on the reliability analysis parameters.The method and device provided by the application are used to solve the problem that the reliability of a single seal cannot be calculated under group test, realize the dual goals of efficient group test and accurate reliability evaluation of single seal, and effectively shorten the test time and reduce the test cost.

Description

Technical Field

[0001] This application relates to the field of sealing technology, and in particular to a graphical estimation data processing method and apparatus for group reliability testing of sealing components. Background Technology

[0002] As a core component of fluid transport and sealing protection equipment, the service life and reliability of seals directly determine the operational stability, safety, and service life of the entire equipment. Once a seal leaks or fails, it can easily lead to equipment failure, media loss, or even safety accidents. Therefore, conducting reliability tests on seals and accurately assessing their reliability level is a necessary step in the research, development, production, and engineering application of seals, and an important prerequisite for ensuring the overall performance of the equipment.

[0003] Currently, there are two main approaches to reliability testing of seals. One approach involves individually assessing the lifespan of each seal, recording its lifespan data after a single seal fails due to leakage, and then statistically analyzing the failure lifespan data of multiple seals to determine reliability. The other approach utilizes the capability of some equipment to connect multiple seals in series, conducting lifespan tests on multiple seals in a group to achieve simultaneous testing. However, both existing testing methods have significant problems and shortcomings. While individual testing ensures the accuracy of reliability calculations, it requires testing each seal until failure, resulting in long testing cycles, high manpower and material costs, and high testing costs. While grouped series testing significantly improves testing efficiency and shortens testing time, it is limited by testing rules and can only record one lifespan data point per group, making it impossible to calculate the reliability of a single seal based on this data. This makes it difficult to balance testing efficiency with the accuracy of reliability assessment.

[0004] Therefore, there is an urgent need for a method to solve the problem of not being able to calculate the reliability of individual seals under group testing, so as to achieve the dual goals of efficient group testing and accurate reliability assessment of individual seals, while effectively shortening the test time and reducing the test cost. Summary of the Invention

[0005] In view of this, this application provides a graphical estimation data processing method and apparatus for group reliability testing of seals, which solves the problem that the reliability of a single seal cannot be calculated under group testing, and achieves the dual goals of efficient group testing and accurate reliability assessment of a single seal, while effectively shortening the test time and reducing the test cost.

[0006] Specifically, this application is implemented through the following technical solution:

[0007] The first aspect of this application provides a graphical estimation data processing method for group reliability testing of sealing components, the method comprising:

[0008] Each group of seals is treated as a sample. Based on the distribution characteristics of the seal failure life data, the distribution parameters are calculated by graphical estimation, and the basic fitting curve is plotted.

[0009] By combining the number of components in the group test of the sealing components and the test stopping rules, the basic fitting curve is shifted, and the reliability estimation diagram corresponding to the group test is plotted.

[0010] Based on the reliability estimation diagram, the reliability analysis parameters for the seal life test are calculated, and the reliability of the seal is evaluated based on the reliability analysis parameters.

[0011] A second aspect of this application provides a graphical estimation data processing apparatus for group reliability testing of seals, the apparatus comprising a calculation module, a plotting module, and an evaluation module;

[0012] The calculation module is used to treat each group of seals as a sample, calculate the distribution parameters through graphical estimation based on the distribution characteristics of the seal failure life data, and draw the basic fitting curve.

[0013] The plotting module is used to combine the number of components in the group test of the sealing components and the test stopping rules to offset the basic fitting curve and plot the reliability estimation diagram corresponding to the group test.

[0014] The evaluation module is used to calculate the reliability analysis parameters of the seal life test based on the reliability estimation diagram, and to evaluate the reliability of the seal based on the reliability analysis parameters.

[0015] The graphical estimation data processing method and apparatus for group reliability testing of seals provided in this application first treats each group of seals as a sample. Based on the distribution characteristics of the seal failure life data, the distribution parameters are calculated by graphical estimation and a basic fitting curve is plotted. This step transforms the group-level failure life data that can only be obtained in group tests into a quantifiable and analyzable group-level life-probability law curve, providing a stable and reliable data foundation for subsequent reliability calculation of individual seals. This truly transforms the group test data that was originally unusable into an effective analytical basis. Secondly, by combining the number of components within a group and the test stopping rules in the group test, the basic fitting curve is offset to obtain a reliability estimation map. Based on the group test conditions, the group-level law curve is converted into a single-component law curve, which directly solves the industry pain point that traditional group tests can only measure efficiency and cannot calculate reliability. Without changing the test equipment, increasing the test operations, or extending the test time, the failure law at the group level is accurately mapped to the failure law of a single seal. This not only inherits the advantages of group tests in simultaneously testing multiple seals and high test efficiency, but also breaks through the fundamental limitation of the original technology that cannot evaluate the reliability of a single component, thus achieving the unity of test method and evaluation goal. Finally, reliability analysis parameters are calculated based on the reliability estimation map, and the reliability assessment of the seal is completed. The reliability estimation map obtained from the previous two steps directly supports the engineering reliability assessment, eliminating the need to test each seal individually until failure. This achieves a closed-loop process of "group testing for data, graph estimation for fitting, offset transformation for individual components, and parameter calculation for assessment," truly achieving the dual goals of efficient group testing and accurate reliability assessment of individual seals. While significantly shortening the test cycle, reducing test sample consumption, and lowering the overall test cost, it can also output quantitative reliability indicators that can be directly used for product design, inspection, and acceptance. Attached Figure Description

[0016] Figure 1 A flowchart of a graphical estimation data processing method for a group reliability test of seals provided in Embodiment 1 of this application;

[0017] Figure 2 This is a schematic diagram illustrating the two curves shown in this application;

[0018] Figure 3 This is a schematic diagram of the structure of the graphical estimation data processing device for the group reliability test of seals provided in Embodiment 2 of this application. Detailed Implementation

[0019] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application.

[0020] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used herein are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

[0021] It should be understood that although the terms first, second, third, etc., may be used in this application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0022] The following specific embodiments are given to illustrate the technical solution of this application in detail.

[0023] Figure 1 This is a flowchart of the graphical estimation data processing method for the group reliability test of seals provided in Embodiment 1 of this application. Please refer to... Figure 1 The method provided in this embodiment may include:

[0024] S101. Treat each group of seals as a sample. Based on the distribution characteristics of the seal failure life data, calculate the distribution parameters through graphical estimation and plot the basic fitting curve.

[0025] The distribution characteristics of the failure life data of the seal are Weibull distribution.

[0026] It should be noted that before plotting the basic fitting curve, the test seals need to be classified and grouped twice. First, they should be initially divided according to the factory standard failure life distribution curve, and then further subdivided according to the usage environment parameters. This will make the life development patterns of the seals in the same group relatively consistent, which will not only improve the success rate and accuracy of subsequent Weibull distribution fitting, but also avoid the omission of important life characteristic information due to the removal of deviation data.

[0027] Optionally, the factory standard failure life distribution curve of the seals to be tested is retrieved first. This curve represents the baseline life pattern tested at the factory, and its core components include the Weibull distribution shape parameter m, scale parameter η (characteristic life), position parameter γ, as well as key parameters such as the factory-calibrated rated fatigue life and the baseline life corresponding to 50% / 90% reliability. These parameters serve as the core criterion for the first grouping, ensuring that the seals after initial grouping maintain homogeneity in their inherent factory life characteristics. Furthermore, based on the deviation range of the core parameters of the seal's factory standard failure life distribution curve, all seals to be tested are initially grouped in batches to ensure that the inherent factory life patterns of seals within the same initial group are highly consistent. This avoids deviations in life characteristics caused by minor differences in factory batches, manufacturing processes, and raw materials, laying the foundation for subsequent environmental parameter subdivision. The core judgment thresholds are the shape parameter m (deviation ±5%), dimensional parameter η (deviation ±3%), and 90% reliability baseline life (deviation ±4%) of the factory standard Weibull distribution. Auxiliary parameters such as the factory-calibrated rated fatigue life, material hardness, and sealing surface accuracy are also referenced. The factory test parameters of all seals to be tested are compared with the standard parameters. Seals with parameter deviations within the above thresholds are grouped into the same initial group, and those with deviations exceeding the thresholds are grouped into a separate initial group. This ensures that each initial group is a batch with homogeneous factory life characteristics, thus forming several initial groups of seals with consistent factory life characteristics.

[0028] Furthermore, within each initial group, clustering is based on the actual service environment parameters of the seals. Through multi-dimensional environmental parameter quantification, the initial groups are further subdivided into several smaller test groups to ensure that the seals within each group face consistent environmental stresses in actual use. The core clustering environmental parameters (quantifiable and detectable, covering the main service stresses of the seals) include media parameters, operating condition parameters, environmental parameters, and installation parameters. Media parameters include the type of sealing medium (hydraulic oil / water / gas, etc.), medium temperature (°C), medium pressure (MPa), and medium impurity content (mg / L); operating condition parameters include the seal's reciprocating / rotating frequency (times / min / rpm), working stroke (mm), and sealing surface contact pressure (MPa); environmental parameters include the service environment temperature (°C), ambient humidity (%RH), and environmental corrosion level (1-5); installation parameters include installation interference (mm), mating part surface roughness (Ra), and installation coaxiality (mm).

[0029] In practice, for the seals in each initial group, the K-means clustering algorithm is used to perform cluster analysis on all the above-mentioned quantified environmental parameters. The clustering threshold is "the coefficient of variation of each environmental parameter in the same group is ≤8%", and seals with highly similar environmental parameters are grouped into the same small test group. At the same time, the number of seals in each small test group (e.g., 4 pieces / group, 5 pieces / group) is determined according to the equipment load-bearing capacity of the group test, and the final test grouping is completed.

[0030] The method provided in this embodiment, after two groupings, ensures that the seals within each test group exhibit no deviations in their factory life characteristics or stress differences due to the usage environment. In the grouped tests, the failure sequence and life decay patterns of the seals within each group are highly synchronized. The first failure life data collected at the group level, collected when "the test stops if any seal leaks," accurately reflects the inherent lifespan of this type of seal under the corresponding environment, avoiding the problem of excessively high dispersion of group-level data due to sample heterogeneity. The homogeneous group-level failure life data significantly reduces the dispersion of data points in the Weibull distribution plot estimation and fitting. When using the least squares method for fitting, the data points fit the fitted line more closely, enabling rapid and accurate solution of the Weibull distribution parameters and significantly improving the basic fitting curve. The success rate of line plotting is improved without the need for repeated removal of outlier data, and fitting efficiency and accuracy are improved simultaneously. The two groupings are hierarchical homogeneous groupings, rather than simple "removal of deviation data". All seals to be tested are classified into corresponding groups according to their own factory characteristics and environmental parameters. Group tests are carried out in different groups and basic fitting curves are plotted separately. This not only preserves the differences in life characteristics of seals with different factory characteristics and different usage environments, but also makes each group of data have fitting value. It avoids the omission of special life patterns of seals under different operating conditions and different batches due to the removal of so-called "deviation data" (such as accelerated life decay under high temperature conditions, early failure under high impurity media, etc.). This allows the subsequent reliability assessment to cover the full-scenario service characteristics of the seals.

[0031] Specifically, failure life data refers to the first failure life data recorded for each group of seals in a group reliability test, according to the test termination rules. This data represents the service life of each seal from the start of the test until the first seal in the group experiences leakage failure (e.g., 8000, 11000, 15500, 17000, 22000, 24000, 30000, 38000 cycles, etc., representing the number of reciprocating fatigue cycles). Distribution characteristics refer to the Weibull distribution pattern exhibited by the seal failure life data at the statistical level. This is an inherent statistical characteristic of seal failure life and a common feature of seal life data.

[0032] Furthermore, the distribution parameters refer to the characteristic parameters of the Weibull distribution obtained by fitting the failure life data of the group tests to the Weibull distribution using graphical estimation methods. These parameters are key values ​​characterizing the Weibull distribution law of the seal failure life. The core parameters of the Weibull distribution include shape parameters, scale parameters (characteristic life), and location parameters. These parameters together determine the shape, position, and trend of the Weibull distribution curve, quantitatively reflecting the statistical law of the seal failure life, and also serving as the core numerical support for plotting the basic fitting curve.

[0033] The basic fitting curve refers to the Weibull distribution fitting line obtained by treating each group of seals as a sample, plotting all group-level failure life data obtained from group tests, combined with the corresponding group-level failure probabilities, on a Weibull distribution coordinate graph, and then fitting the graph using the least squares method. This curve directly reflects the correspondence between group-level failure life and group-level failure probability under group tests, and only characterizes the life-probability law at the group level.

[0034] In specific implementation, based on the distribution characteristics of the seal failure life data, the distribution parameters are calculated through graphical estimation, and a basic fitting curve is plotted. This includes: sorting the failure life data of each group obtained from the group test in ascending order; calculating the corresponding group-level failure probability for each sorted group life data; plotting each group life data and its corresponding group-level failure probability as data points on a Weibull distribution coordinate graph; and fitting the data points using the least squares method to obtain the basic fitting curve.

[0035] Calculate the group-level failure probability for each sorted group lifetime data, including: recording the total number of groups in the grouped experiment as the total number of samples; labeling the sorted group lifetime data with serial numbers in sequence, using the result of subtracting the first correction coefficient of the median rank from the serial number as the numerator, and using the result of adding the second correction coefficient of the median rank to the total number of samples as the denominator; and obtaining the group-level failure probability corresponding to each group lifetime data by dividing the numerator by the denominator.

[0036] Specifically, the failure life data of seals from all test groups in the group reliability test are collected. All group-level failure life data are arranged sequentially from smallest to largest value, forming an ordered sequence of group life data. Further, the total number of test groups in this group test is defined as the total sample size. The sorted group life data are sequentially numbered from 1, with each group having a unique number. The median rank first correction factor is pre-set to 0.3, and the median rank second correction factor is set to 0.4. For each group life data with a marked number, a separate calculation is performed: the median rank first correction factor is subtracted from the group number to obtain the numerator of the probability calculation; the median rank second correction factor is added to the total sample size to obtain the denominator of the probability calculation; the numerator is divided by the denominator to obtain the group-level failure probability corresponding to that group life data. The group-level failure probabilities corresponding to all sorted group life data are calculated sequentially using the above method to obtain the unique group-level failure probability for each group life data. Prepare a Weibull distribution-specific coordinate plot, with the horizontal axis representing the group-level failure lifetime and the vertical axis representing the group-level failure probability. Mark all data points on the Weibull distribution coordinate plot, ensuring that the horizontal and vertical coordinates of each data point precisely correspond to the lifetime data and failure probability of the corresponding group. Use the least squares method to perform linear fitting on all marked data points on the Weibull distribution coordinate plot, generating a fitted straight line based on the distribution trend of the data points. This fitted straight line is the basic fitted curve. Simultaneously, during the calculation of this linear fitting, the distribution parameters of the Weibull distribution are solved.

[0037] Optionally, after completing the group-level failure probability calculation and marking the data points on the Weibull distribution coordinate graph, if the number of groups obtained after the experiment is too large, resulting in scattered data point distribution on the coordinate graph and easy deviation in direct linear fitting, all data points can be clustered and merged first, and the basic fitting curve can be transformed from a single straight line into a broken line fitting. The inflection point of the broken line is determined based on the density distribution characteristics of data points within a single category after clustering. When the density of data points changes abruptly, a broken line inflection point is added.

[0038] In practical implementation, the horizontal axis (group-level failure lifetime) of the Weibull distribution coordinate graph is used as the core dimension, combined with the vertical axis (group-level failure probability). All group-level lifetime-failure probability data points on the coordinate graph are initially clustered according to lifetime value range + probability change gradient. A density clustering algorithm is used, with clustering conditions of "neighborhood density ≥ preset threshold, lifetime range deviation ≤ 10%, probability deviation ≤ 5%". Multiple data points with similar characteristics are clustered into one data category. Each category represents a set of group-level lifetime data with highly consistent characteristics. The mean lifetime, mean probability, and density distribution characteristics of data points within each category are retained. All clustered categories are arranged sequentially in ascending order of lifetime value on the horizontal axis. The rate of change in data point density between adjacent categories is calculated. If the rate of change in data point density of a later category relative to the previous category is ≥ 50%, it is considered a density abrupt change, and the location of this density abrupt change is the inflection point of the line fitting. If the rate of change in density is < 50%, it is considered a density plateau, and no inflection point is needed; adjacent categories can be further merged. Based on the clustered categories, within a stable density interval, the mean data points of all categories within that interval are linearly fitted using the least squares method, forming a straight line segment of a polygonal line. At the inflection point of a sudden density change, the linearly fitted straight line segment of the next density interval is connected. Finally, multiple consecutive straight line segments form a polygonal line that fits the distribution characteristics of the data points. This polygonal line is the basic fitting curve after clustering and merging. During the fitting process, the Weibull distribution parameters corresponding to each straight line segment are solved simultaneously to preserve the distribution characteristics of each segment.

[0039] The method provided in this embodiment has already undergone two stratified groupings in the early stage, ensuring that the samples in each experimental group have homogeneous characteristics. The fully collected group-level failure life data and the calculated failure probabilities completely reflect the life development patterns of seals under different manufacturing characteristics and usage environments. This clustering and merging only classifies and integrates data points with similar characteristics, without removing any experimental data. The life characteristics of all original data are preserved through the mean and density distribution of the cluster categories, ensuring that the key patterns of seals at different life stages and probability intervals are not lost due to the merging operation, and completely avoiding the problem of missing important information caused by data removal. To address the problem of scattered data points caused by too many groups, density clustering is used to merge multiple data points with similar characteristics into one category, significantly reducing the number of basic data units required for fitting. This allows the originally scattered massive data points to form several distinct cluster intervals, solving the problem of large fitting deviation and low success rate caused by excessively high data point dispersion in single linear fitting, simplifying the subsequent fitting calculation process, and improving the efficiency of drawing the basic fitting curve. Compared to the strong constraints imposed by single-line fitting on data point distribution, piecewise linear fitting incorporates inflection points based on abrupt changes in data point density. This ensures that each line segment closely matches the true distribution trend of data points within its corresponding density range, accurately reflecting the failure probability changes of seals at different life stages. For example, the rapid increase in probability during the early failure stage, the gradual change in probability during the mid-term stable failure stage, and the sharp increase in probability during the late failure stage can all be reflected by the slope of different line segments. The fitting results more closely match the actual data characteristics of the group experiments, providing a more accurate basis for subsequent curve offset processing. Furthermore, using abrupt changes in data point density as the criterion for determining inflection points ensures that there are significant differences in the life characteristics of seals corresponding to different line segments, avoiding meaningless line segmentation. This allows piecewise linear fitting to retain the detailed features of the data without excessive curve complexity due to over-segmentation, achieving a balance between data simplification and feature preservation. This makes subsequent group-level pattern analysis based on the basic fitted curve clearer and more targeted.

[0040] For example, Figure 2 This is a schematic diagram illustrating the two curves shown in this application. Please refer to... Figure 2 , Figure 2 Line A in the figure is the basic fitted curve. The failure probability of point C, calculated using the method described above, is 44%, which is 50% after fitting.

[0041] S102. Based on the number of components in the group test and the test termination rules, the basic fitting curve is shifted, and a reliability estimation diagram corresponding to the group test is drawn.

[0042] The test stop rule includes stopping the test and recording the current lifespan data when any set of seals in each group leaks.

[0043] Specifically, the test termination rule refers to the standardized test termination criteria followed during group testing. A predetermined number of seals are grouped together for synchronous life testing. If any set of seals in the group fails due to leakage during the test, all tests for that group are immediately stopped, and the service life data of the corresponding seals in that group is recorded. This data is the group-level failure life data for that group of seals. All test groups follow this rule. The reliability estimation graph refers to the fitted curve (offset fitted curve) representing the relationship between the failure life and failure probability of a single seal under group testing. This graph uses the service life of the seal as the horizontal axis and the failure probability of a single seal as the vertical axis, directly reflecting the failure probability pattern of a single seal at different service lives.

[0044] It should be noted that the basic fitting curve is obtained by fitting group-level failure life data and group-level failure probability. It only reflects the life-failure probability correspondence at the group level, representing the probabilistic law that "at least one of the seals in a group fails," not the failure law of a single seal. Therefore, it cannot be directly used for the reliability assessment of a single seal. The number of parts within a group in the group test and the "stop if a leak occurs" test cessation rule are the core factors causing the difference between the group-level failure law and the failure law of a single seal. These two key test conditions must be combined, and curve offset processing must be used to convert the group-level life-probability law into the life-probability law of a single seal in order to achieve the derivation from group-level test data to the reliability assessment of a single seal. That is, the essence of offset processing is to eliminate the influence of group test conditions on the failure probability law and complete the accurate conversion from group-level law to single-part law.

[0045] In specific implementation, the basic fitting curve is offset to draw a reliability estimation map corresponding to the group test, including: selecting a reference point corresponding to any reference group level failure probability on the basic fitting curve; converting the reference group level failure probability into the corresponding single seal failure probability based on the number of seals in each group of the group test; determining the target point corresponding to the single seal failure probability in the coordinate graph; drawing an offset fitting curve parallel to the basic fitting curve to obtain the reliability estimation map corresponding to the group test.

[0046] Based on the number of seals in each group of group tests, the reference group-level failure probability is converted into the corresponding individual seal failure probability, including: recording the number of seals in each group as the group quantity; subtracting the reference group-level failure probability from the number 1 to obtain the probability value that all seals in the group have not failed; raising the probability value to the power of the group quantity to obtain the probability value that an individual seal has not failed; and subtracting the probability value that an individual seal has not failed from the number 1 to obtain the individual seal failure probability corresponding to the reference group-level failure probability.

[0047] Specifically, a reference group-level failure probability is arbitrarily selected on the basic fitting curve, and a unique coordinate point corresponding to the reference group-level failure probability is determined on the basic fitting curve. This coordinate point is used as the reference point for this curve offset processing, and the group-level failure lifetime value and the reference group-level failure probability value corresponding to the reference point are recorded. Furthermore, in this group reliability test, the actual number of seals in each group was counted, and this number was defined as the group quantity, with this parameter fixed. Using 1 as the minuend, the selected reference group-level failure probability was subtracted. The result of this subtraction operation is the probability that all seals in the group will not leak or fail within the corresponding lifespan. Using this probability as the base and the determined group quantity as the root, the square root of this probability value was calculated. The result is the probability that a single seal will not leak or fail within the corresponding lifespan. Using 1 as the minuend, the probability of a single seal not failing was subtracted. The final result of this subtraction operation is the single seal failure probability value corresponding one-to-one with the selected reference group-level failure probability. Using the group-level failure lifespan value corresponding to the reference point as the horizontal axis and the calculated single seal failure probability value as the vertical axis, a unique coordinate point was determined based on these horizontal and vertical axis values ​​on the same Weibull distribution coordinate graph. This coordinate point was used as the target point for this curve offset processing. Using the defined target point as a reference, draw a straight line on the Weibull distribution coordinate graph that passes through the target point and is parallel to each other with the slope of the original base fitting curve. This straight line is the offset fitting curve after the offset processing.

[0048] For example, the calculation process for the failure probability of a single seal can be expressed as:

[0049] F = 1 - (1 - P)^n;

[0050] Where F is the probability of failure of a single seal; P is the probability of failure of the reference group; and n is the number of seals in each group.

[0051] by Figure 2 Taking point C as a reference point for illustration, the failure probability of the reference group corresponding to the reference point is 50%. According to the calculation formula, the failure probability of a single seal corresponding to point C is 15.9%. A line parallel to the vertical axis is drawn through point C, intersecting the line parallel to the horizontal axis with a failure probability of 15.9% at point M. A line B parallel to line A is drawn through point M. Line B is the offset fitting curve.

[0052] The method provided in this embodiment uses N piecewise straight lines as the basic fitting curve after clustering and merging, and the offset fitting curves correspond to the same number of parallel piecewise straight lines, which together constitute a reliability estimation map. This multi-round processing is not a redundant operation, but rather achieves a precise mapping from group-level lifetime-probability characteristics to individual-item patterns through piecewise offsetting. This significantly reduces the amount of original, scattered experimental data while continuously enriching effective information. From isolated group-level lifetime values, lifetime-probability correlations and staged failure patterns are gradually superimposed, ultimately completing the core upgrade from group-level to individual-item information, while preserving density mutations and grouping. Key features such as probability transformation enable data subtraction and information addition. Through steps such as grouping, clustering, fitting, and shifting, the core is to address the pain points of group experiments, such as sample heterogeneity, data fragmentation, fitting bias, and inability to derive the reliability of individual components. Ultimately, while inheriting the advantages of high efficiency and low cost of group experiments, it breaks through technical limitations to form an estimation map that can accurately characterize the reliability law of individual seals, achieving the dual goals of efficient group experiments and accurate evaluation of individual components. The absence of any step will lead to fitting distortion or failure of law transformation, making it impossible to obtain usable evaluation conclusions.

[0053] S103. Based on the reliability estimation diagram, calculate the reliability analysis parameters for the seal life test, and evaluate the reliability of the seal based on the reliability analysis parameters.

[0054] Specifically, reliability analysis parameters refer to various characteristic values ​​that can quantify the reliability level of a single seal. The core parameter is the seal fatigue life corresponding to a preset reliability level, and it also includes related derived parameters that reflect the seal failure patterns. The core reliability analysis parameter is the seal fatigue life at a specified reliability level. This involves pre-setting a target reliability level for the seal, querying the failure probability of a single seal corresponding to that target reliability level based on a reliability estimation chart, and then determining the seal service life value corresponding to that failure probability on the coordinate graph. This service life value is the core reliability analysis parameter, directly reflecting the actual service capability of the seal under the specified reliability requirements. Derived reliability analysis parameters are parameters derived from the reliability estimation chart that reflect the overall failure pattern of the seal, including seal service life data corresponding to different failure probabilities and quantitative values ​​of the failure probability change trend of the seal throughout its entire life cycle.

[0055] In specific implementation, based on the reliability estimation map, the reliability analysis parameters of the seal life test are calculated, including: in the reliability estimation map, determining the corresponding failure probability of a single seal according to the preset seal reliability; querying the seal life data corresponding to the failure probability in the coordinate map, and using the life data as the fatigue life of the seal under the preset seal reliability to obtain the reliability analysis parameters.

[0056] Optionally, the preset seal reliability includes 90%.

[0057] Specifically, a pre-set reliability value for the seal to be evaluated (e.g., 90%) is used. Based on the relationship between the seal failure probability and the sum of the seal reliability values ​​(1), the failure probability of a single seal corresponding to this pre-set reliability is calculated; for example, when the pre-set reliability is 90%, the corresponding failure probability of a single seal is 10%. In the completed reliability estimation graph, the vertical axis is designated as the scale area for "single seal failure probability," and the vertical axis scale line corresponding to the calculated failure probability value is located. Extending horizontally along the above vertical axis scale line, the intersection point of this scale line and the offset fitting curve in the reliability estimation graph is determined; extending vertically downward from this intersection point to the horizontal axis (seal life value axis), the corresponding value on the horizontal axis is read. This value is the seal life data under this failure probability. The read seal life data is defined as the fatigue life of the seal under this pre-set reliability, and this fatigue life value is the core reliability analysis parameter obtained in this calculation; if multiple different seal reliability values ​​are pre-set, the above steps are followed sequentially to obtain multiple sets of reliability analysis parameters corresponding to different reliability values.

[0058] For example, please continue to refer to Figure 2 When the preset reliability of the seal is 90%, the fatigue life of the point with a failure probability of 10% in line B is 15780 cycles.

[0059] The method provided in this embodiment first treats each group of seals as a sample. Based on the distribution characteristics of the seal failure life data, the distribution parameters are calculated through graphical estimation, and a basic fitting curve is plotted. This step transforms the group-level failure life data, which can only be obtained in group tests, into a quantifiable and analyzable group-level life-probability law curve, providing a stable and reliable data foundation for subsequent reliability calculations of individual seals. This truly transforms group test data, which was originally unusable, into effective analytical basis. Second, the basic fitting curve is offset by combining the number of parts within the group and the test stopping rules of the group test to obtain a reliability estimation map. Based on the group test conditions, the group-level law curve is converted into a single-part law curve, directly solving the industry pain point that traditional group tests can only measure efficiency and cannot calculate reliability. Without changing the test equipment, increasing test operations, or extending the test time, the failure law at the group level is accurately mapped to the failure law of a single seal. This inherits the advantages of group tests, such as simultaneous testing of multiple seals and high test efficiency, while breaking through the fundamental limitation of the original technology that cannot evaluate the reliability of a single part, thus achieving the unity of test method and evaluation objective. Finally, reliability analysis parameters are calculated based on the reliability estimation map, and the reliability assessment of the seal is completed. The reliability estimation map obtained from the previous two steps directly supports the engineering reliability assessment, eliminating the need to test each seal individually until failure. This achieves a closed-loop process of "group testing for data, graph estimation for fitting, offset transformation for individual components, and parameter calculation for assessment," truly achieving the dual goals of efficient group testing and accurate reliability assessment of individual seals. While significantly shortening the test cycle, reducing test sample consumption, and lowering the overall test cost, it can also output quantitative reliability indicators that can be directly used for product design, inspection, and acceptance.

[0060] Corresponding to the aforementioned embodiment of a graphical estimation data processing method for a group reliability test of seals, this application also provides an embodiment of a graphical estimation data processing apparatus for a group reliability test of seals.

[0061] Figure 3 This is a schematic diagram of the graphical estimation data processing device for the group reliability test of seals provided in Embodiment 2 of this application. Please refer to... Figure 3 The apparatus provided in this embodiment includes a calculation module 210, a drawing module 220, and an evaluation module 230;

[0062] The calculation module 210 is used to treat each group of seals as a sample, calculate the distribution parameters through graphical estimation based on the distribution characteristics of the seal failure life data, and draw the basic fitting curve.

[0063] The plotting module 220 is used to combine the number of components in the group test of the sealing component group test and the test stopping rules to offset the basic fitting curve and plot the reliability estimation diagram corresponding to the group test.

[0064] The evaluation module 230 is used to calculate the reliability analysis parameters of the seal life test based on the reliability estimation diagram, and to evaluate the reliability of the seal based on the reliability analysis parameters.

[0065] The apparatus of this embodiment can be used to perform... Figure 1 The steps of the method embodiment shown are similar in principle and process, and will not be repeated here.

[0066] The specific implementation process of the functions and roles of each unit in the above device can be found in the implementation process of the corresponding steps in the above method, and will not be repeated here.

[0067] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this application according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0068] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for processing graphical estimation data in a group reliability test of sealing components, characterized in that, The method includes: Each group of seals is treated as a sample. Based on the distribution characteristics of the seal failure life data, the distribution parameters are calculated by graphical estimation, and the basic fitting curve is plotted. Based on the number of components within a group and the test termination rules in the group test of the sealing components, the basic fitting curve is shifted to generate a reliability estimation diagram corresponding to the group test. The test termination rules refer to the test termination judgment criteria uniformly followed during the group test. A preset number of sealing components are used as a group to carry out synchronous life tests. If any set of sealing components in the target group fails due to leakage during the test, all tests of the target group are immediately stopped, and the service life data of the sealing components corresponding to the target group test at the current moment is recorded. The service life data of the sealing components is the group-level failure life data of the sealing components in the target group. Based on the reliability estimation diagram, the reliability analysis parameters of the seal life test are calculated, and the reliability of the seal is evaluated based on the reliability analysis parameters. The baseline fitting curve is shifted, and a reliability estimation plot corresponding to the group of experiments is drawn, including: Select a reference point corresponding to the failure probability of any reference group on the basic fitting curve; Based on the number of seals in each group of group tests, the reference group-level failure probability is converted into the corresponding individual seal failure probability. Determine the target point corresponding to the failure probability of the individual seal in the coordinate graph, draw an offset fitting curve parallel to the basic fitting curve, and obtain the reliability estimation map corresponding to the group test.

2. The graphical estimation data processing method for group reliability testing of seals according to claim 1, characterized in that, Based on the distribution characteristics of the seal failure life data, the distribution parameters are calculated through graphical estimation, and the basic fitting curve is plotted, including: The failure life data of each group obtained from the group test are sorted in ascending order; Calculate the group-level failure probability for each sorted group lifetime data; The lifetime data of each group and the corresponding group-level failure probability are used as data points and plotted on a Weibull distribution coordinate graph. The data points are fitted using the least squares method to obtain the basic fitted curve.

3. The graphical estimation data processing method for group reliability testing of seals according to claim 2, characterized in that, Calculate the group-level failure probability for each sorted group lifetime data, including: The total number of groups in a grouped experiment is recorded as the total sample size. The sorted group lifetime data are sequentially labeled with serial numbers. The result of subtracting the first correction coefficient of the median rank from the serial number is used as the numerator, and the result of adding the second correction coefficient of the median rank to the total number of samples is used as the denominator. The group-level failure probability corresponding to each set of lifetime data is obtained by dividing the numerator by the denominator.

4. The graphical estimation data processing method for group reliability testing of seals according to claim 1, characterized in that, Based on the number of seals in each group of group tests, the reference group-level failure probability is converted into the corresponding individual seal failure probability, including: The number of seals in each group is recorded as the group quantity. The probability that all seals in the group are not failed is obtained by subtracting the reference group-level failure probability from the number 1. The probability value is raised to the power of the group number to obtain the probability value that a single seal has not failed; Subtracting the probability value of a single seal not failing from the number 1 yields the probability of a single seal failing, which corresponds to the reference group-level failure probability.

5. The graphical estimation data processing method for group reliability testing of seals according to claim 1, characterized in that, Based on the reliability estimation diagram, the reliability analysis parameters for the seal life test are calculated, including: In the reliability estimation diagram, the failure probability of a single seal is determined according to the preset seal reliability. Query the seal life data corresponding to the failure probability in the coordinate graph, and use the life data as the fatigue life of the seal under the preset seal reliability to obtain reliability analysis parameters.

6. The graphical estimation data processing method for group reliability testing of seals according to claim 5, characterized in that, The preset reliability of the seal includes 90%.

7. The graphical estimation data processing method for group reliability testing of seals according to claim 1, characterized in that, The test stop rule includes stopping the test and recording the current lifespan data when any set of seals in each group leaks.

8. The graphical estimation data processing method for group reliability testing of seals according to claim 1, characterized in that, The distribution characteristics of the failure life data of the seal are Weibull distribution.

9. A graphical estimation data processing device for group reliability testing of seals, characterized in that, The device includes a calculation module, a drawing module, and an evaluation module; The calculation module is used to treat each group of seals as a sample, calculate the distribution parameters through graphical estimation based on the distribution characteristics of the seal failure life data, and draw the basic fitting curve. The plotting module is used to offset the basic fitting curve by combining the number of components in the group test of the sealing components and the test stop rules, and plot the reliability estimation map corresponding to the group test. The test stop rules refer to the test termination judgment criteria uniformly followed in the group test. A preset number of sealing components are used as a group to carry out synchronous life tests. If any set of sealing components in the target group fails due to leakage during the test, all tests of the target group are immediately stopped, and the service life data of the sealing components corresponding to the target group test at the current moment is recorded. The service life data of the sealing components is the group-level failure life data of the sealing components in the target group. The evaluation module is used to calculate the reliability analysis parameters of the seal life test based on the reliability estimation diagram, and to perform a reliability evaluation of the seal based on the reliability analysis parameters. The baseline fitting curve is shifted, and a reliability estimation plot corresponding to the group of experiments is drawn, including: Select a reference point corresponding to the failure probability of any reference group on the basic fitting curve; Based on the number of seals in each group of group tests, the reference group-level failure probability is converted into the corresponding individual seal failure probability. Determine the target point corresponding to the failure probability of the individual seal in the coordinate graph, draw an offset fitting curve parallel to the basic fitting curve, and obtain the reliability estimation map corresponding to the group test.

Images