Reliability data mining and evaluation method of lightning protection components based on HALT test

Abstract

The present invention discloses a method for mining and evaluating the reliability data of lightning protection components based on HALT tests, which relates to the field of environmental and reliability testing technology and includes the following specific steps: S1, conducting a preliminary HALT test of lightning protection components using a single step stress; S2, exploring the multi-stress coupling effect based on a DOE test design; S3, calculating the acceleration factor of a single stress and a combined stress; and S4, data conversion and reliability evaluation based on the acceleration factor. This method for mining and evaluating the reliability data of lightning protection components based on HALT tests considers the influence of complex and combined environmental stresses, provides a more reasonable calculation method for calculating the combined stress acceleration factor, and improves the credibility of the evaluation data. The data conversion method helps convert HALT test data into a data form that can be used for reliability evaluation, thereby obtaining a reliability evaluation result for the lightning protection components and improving the data's usability.

Description

Technical Field

[0001] The present invention relates to the technical field of environment and reliability testing, and in particular to a reliability data mining and evaluation method for lightning protection components based on HALT testing. Background Art

[0002] In modern industrial production and technological research and development, product reliability is a key factor in determining market competitiveness, user satisfaction, and the long-term development of a company. However, in practice, due to various cost constraints, such as time, financial, and resource costs, companies often face the challenge of collecting sufficient test data to support evaluation results when conducting reliability tests. This is especially true for products with extremely high reliability, such as aerospace equipment and medical devices. Qualification tests for their high MTBF (mean time between failures) values ​​often require significant time to complete, and this extended wait may exceed a company's budget or time constraints. Furthermore, high product costs limit the number of test products a company can provide, making comprehensive and in-depth lifespan assessments difficult. Consequently, the reduced amount of data directly reduces the credibility of product reliability evaluation results.

[0003] To meet the above challenges, the present invention optimizes the HALT test process and fully mines the available data of the HALT test to increase the test data information that can be used for reliability evaluation, thereby expanding the scope of the HALT test. For one of the main products, the lightning protection component, a lightning protection component reliability data mining and evaluation method based on the HALT test is proposed.

[0004] Ensuring the reliability of lightning protection components is crucial in their development and lifecycle management. Reliability assessment involves systematically analyzing whether components meet their intended functional requirements under specific operating environments or conditions. Performance degradation and failure data are key elements of this evaluation, providing valuable information for understanding product robustness, predicting service life, and optimizing design.

[0005] However, traditional environmental tests and reliability tests are often limited by time and cost, making it difficult to obtain sufficient performance degradation data and failure data within the original test time. Such tests are usually designed to simulate the performance of lightning protection components under normal operating conditions, but often ignore extreme conditions or potential failure modes. In contrast, HALT tests accelerate the performance degradation process of lightning protection components by introducing stress levels higher than the design specifications, such as high voltage, extreme temperature changes, and strong vibrations, thereby observing more failure phenomena and performance degradation signs in a shorter period of time. During the implementation of the HALT test, the samples are closely monitored and the changes in performance parameters, the occurrence of failure modes, and other information related to the reliability of the lightning protection components are recorded.

[0006] In view of the unique advantages of the HALT test, the present invention focuses on effective data mining of the HALT test, aiming to extract valuable information from the data collected in the test; by optimizing the HALT test process, the efficiency and effect of the test are improved, and the accuracy and credibility of the reliability evaluation of the lightning protection components are increased. By optimizing the HAIT test process and the mining of effective data, it is possible not only to obtain the performance degradation and failure data of the lightning protection components within limited time and resources, but also to propose a more scientific and accurate lightning protection component reliability evaluation method, providing strong support for product design, production and maintenance.

[0007] Prior art related to the present invention

[0008] (1) Currently, the HALT test has been carried out to a relatively mature stage, but the laboratories and related units are only concerned with its basic purpose, which is to quickly determine the stress limit of the test sample and expose potential defects by strengthening the stress stimulation effect. The existing test process cannot fully tap into the available and effective test data for evaluation;

[0009] (2) The current acceleration factor calculation method mainly relies on the acceleration equation calculation under single stress, including the acceleration factor of single vibration acceleration stress, single high temperature stress acceleration factor, single humidity acceleration factor calculation, etc. There are still deficiencies in the calculation of comprehensive acceleration factors. Due to the strong coupling effect under comprehensive stress, it is unreasonable to simply multiply a single acceleration factor.

[0010] (3) The current reliability assessment method mainly relies on probability theory and mathematical statistics to process data. However, due to the particularity of the data form of the HALT test, the existing methods cannot be used. It needs to be processed and converted into effective assessment data before it can be used;

[0011] Therefore, how to optimize the HALT test process, find a more reasonable acceleration factor calculation method, transform and process the HALT test data, fully explore the potential data of the HALT test, and increase the available reliability data are still key issues that need to be solved urgently. Summary of the Invention

[0012] In view of the shortcomings of the existing technology, the present invention provides a lightning protection component reliability data mining and evaluation method based on HALT test, which solves the problems raised in the above background technology.

[0013] To achieve the above objectives, the present invention is implemented through the following technical solutions: a lightning protection component reliability data mining and evaluation method based on HALT test, comprising the following specific steps:

[0014] S1. Single step stress HALT test of lightning protection components is as follows:

[0015] When conducting a reliability assessment of a lightning protection component, it is first necessary to comprehensively collect the design parameters and key performance indicators of the lightning protection component. Design parameters include but are not limited to dimensions, materials, and structural characteristics. Key performance indicators are used to measure the performance of the lightning protection component during operation. Based on the laboratory's capabilities, a single stress step test profile is designed.

[0016] S2: Investigation of multi-stress coupling effects based on DOE experimental design, as follows:

[0017] S201. Screen key performance parameters. Preliminarily screen key performance parameters and understand the changing trends of these performance parameters through a single step stress HALT test. Collect performance data on lightning protection components under different single stresses, screen out key performance parameters with degradation trends, thereby reflecting the response characteristics of lightning protection components under long-term stress, and establish a performance degradation model.

[0018] S202. Design a DOE test plan based on performance degradation. Based on the degradation and reaction rate of the key performance parameters of the lightning protection component under a single stress, determine the stress levels and number of levels for each single stress in the DOE test plan. On this basis, reasonably select a DOE test plan of orthogonal design, uniform design, or single regression orthogonal test.

[0019] This DOE test plan needs to be based on the results of the single step stress HALT test to determine the test time for each level of each stress, to ensure that at each stress level, the degradation data of the key performance of the lightning protection component has at least 3% to 5% degradation, and to determine the test time for each stress level. The test time for different levels of each stress should be consistent, and the time when the minimum level reaches the minimum degradation standard shall prevail.

[0020] ;

[0021] in, is the performance indicator, For initial performance, is the degradation rate, For time;

[0022] S203. Collect DOE test results, perform variance analysis and partial regression coefficient test, conduct tests according to the DOE test design, collect the degradation of key performance parameters of lightning protection components, use variance analysis method to obtain the relationship between stress and key performance parameters, namely the main effect and interaction effect, select the significance level, unify the variance analysis results, and lay the foundation for the calculation of acceleration factors;

[0023] S3. Calculation of acceleration factors for single stress and combined stress is as follows:

[0024] S301, temperature stress acceleration factor, using the Arrhenius model as the temperature stress acceleration model, temperature accelerated stress T s Acceleration factor A under relative normal stress T0 T The calculation formula is:

[0025] ;

[0026] Where, It is the product life characteristics under normal temperature stress; is the product life characteristics under accelerated stress; E a is the activation energy, in eV; k is the Boltzmann constant, k=8.617x×10 -5 eV / K; T0 is the absolute temperature value of normal temperature stress, unit is K; T s is the absolute temperature value of the accelerated temperature stress, in K;

[0027] S302, humidity acceleration factor, using the Peck model as the acceleration model, the humidity stress acceleration factor calculation formula is:

[0028] ;

[0029] RH0 is normal humidity stress; RH S To accelerate humidity stress; n is 2~3;

[0030] S303, vibration acceleration factor, using the inverse power law model as the acceleration model, the vibration stress acceleration factor calculation method is:

[0031] ;

[0032] W0 is the normal vibration stress level; W1 is the accelerated vibration stress level; V0 is the test time under the normal vibration stress level; V1 is the test time under the accelerated vibration stress level;

[0033] S304. Calculation method for acceleration factors based on temperature stress. Under comprehensive environmental stress, the effects of various stresses on products are coupled, classified as mutual independence, mutual inhibition, and mutual promotion. In the independent case, the acceleration factor is the multiplication of the acceleration factors of the individual stresses. However, in the inhibition / promotion case, the magnitude of the mutual promotion and inhibition effects needs to be quantified.

[0034] A single temperature stress acceleration factor is selected as the benchmark, and the acceleration factor calculation formula is obtained by combining the normalized regression coefficient in step 2:

[0035] ;

[0036] The acceleration factors under other single accelerated stress and comprehensive accelerated stress are obtained from the above formula: ;

[0037] S4. Data conversion and reliability assessment based on acceleration factors are as follows:

[0038] Select the test data corresponding to the cut-off stress level to ensure that the lightning protection components meet the basic assumptions of the accelerated test under various stresses, that is, the failure mechanism is consistent;

[0039] The test time before the cutoff stress level is converted. Tests under different stress levels are equivalent to the test time under the same stress level. The stress lower limit value of the HALT test and DOE test is selected. The test time conversion formula for the HALT test is as follows:

[0040] ;

[0041] in, represents the actual test time under the pth stress step, with a total of M steps; represents the acceleration factor of the first stress step based on the temperature accelerated stress; represents the acceleration factor of the p-th step based on temperature accelerated stress; The test time is converted based on the first step acceleration factor;

[0042] After final conversion, the accelerated test data under a certain stress value will be obtained, that is, the reliability evaluation result of the product will be obtained.

[0043] Optionally, the stress step test profile in step S1 is specifically a test method for systematically increasing external stress and observing changes in the performance of lightning protection components; by gradually increasing and changing external stress, the performance and potential failure modes of lightning protection components under different environments and usage conditions are effectively simulated and evaluated.

[0044] Optionally, the stress step test profile specifically includes a low temperature step test, a high temperature step test, a high temperature and high humidity test, and a vibration step test;

[0045] When designing the test profile, the following points should be considered:

[0046] S101. Selection of stress type: According to the working principle of the lightning protection component and the expected application environment, select the stress type: temperature stress, humidity stress, vibration stress;

[0047] S102, Stress level setting: Based on the design parameters and expected application conditions of the lightning protection components, the stress levels at each stage should be reasonably set to ensure that the test is challenging but not too destructive;

[0048] S103. Stress step size: Determine the step size of the stress level increase to ensure that the test is carried out within a controllable range, avoiding excessively rapid stress increases that may cause the component to be unable to withstand, or excessively slow increases that may not fully reveal the performance and reliability of the component;

[0049] S104. Monitoring and recording: During the entire test process, continuously monitor the performance changes and any abnormal conditions of the lightning protection components, and record key data for subsequent analysis and evaluation.

[0050] Optionally, the performance degradation linear degradation model in step S202 is as follows:

[0051] ;

[0052] in, For the Performance indicators for each sample / moment, is the initial performance intercept, is the linear degradation rate, if , which indicates that performance changes over time Linear decrease; if , the performance improves;

[0053] ;

[0054] in, is the logarithmic transformation of time;

[0055] ;

[0056] in, is the logarithmic transformation of the performance, is the power law index, if , which indicates that performance changes over time Power-law decay;

[0057] ;

[0058] in, For exponential transformation, performance changes exponentially with time t;

[0059] ;

[0060] in, is a scale parameter used to control the degradation rate.

[0061] Optionally, the variance analysis method in step S203 is specifically as follows:

[0062] Assume that each factor has p levels and each level has r experiments. The steps of variance analysis are as follows:

[0063] Solving for the total sum of squares of the variation :

[0064] ;

[0065] in The test results of each group; Represents the average of all test results;

[0066] Solving for the sum of squares of between-group variation :

[0067] ;

[0068] in is the average value of the test results at level i;

[0069] Solving for the within-group sum of squares of variation :

[0070] ;

[0071] in is the average value of the test results at level i;

[0072] Find the F test:

[0073] ;

[0074] in represents the sum of squares of the between-group variation The degree of freedom is p-1; represents the sum of squares of within-group variation The degree of freedom is p(r-1);

[0075] Look up the table to get F, compare the results, if , it is significant, otherwise it is not significant; thus, we can compare the influence of various factors on key performance parameters;

[0076] Solve the multiple linear regression equation for the test results:

[0077] ;

[0078] Standardize the coefficients of the multiple linear regression equation:

[0079] ;

[0080] in, is the standardized regression coefficient; is the independent variable The standard deviation of is the standard deviation of the dependent variable y.

[0081] Optionally, the process of obtaining the reliability evaluation result in step S4 is as follows:

[0082] S401, key performance data degradation modeling, fitting the function of degradation data and time to obtain a key performance parameter degradation model;

[0083] S402: Setting a failure threshold and calculating a pseudo-failure life value of the lightning protection component according to a degradation model;

[0084] S403, performing a hypothesis test on the pseudo failure life value of the sample to determine whether it satisfies normal distribution, Wei_bull distribution, or lognormal distribution;

[0085] S404. Calculate reliability and characteristic life according to the corresponding distribution function, taking normal distribution as an example;

[0086] S405. Based on the acceleration factor, calculate the characteristic life value of the product under normal stress.

[0087] The present invention provides a method for mining and evaluating the reliability data of lightning protection components based on HALT test, which has the following beneficial effects:

[0088] This reliability data mining and evaluation method for lightning protection components based on HALT tests integrates the DOE test design method into the HALT test process. Based on the degradation model under a single step stress and the traditional DOE scheme, a test scheme is designed to ensure that key performance parameters have a certain degradation amount, helping to explore the coupling effect between each single stress on the key performance data of the lightning protection component; considering the influence of complex comprehensive environmental stress, a more reasonable calculation method is provided for the calculation of the comprehensive stress acceleration factor, thereby improving the credibility of the evaluation data; the data conversion method helps to convert the HALT test data into a data form that can be used for reliability evaluation, obtain the reliability evaluation results of the lightning protection component, and improve the availability of the data.

[0089] By integrating the HALT test and DOE test design methods, the HALT test process is optimized. After the single step stress test, a comprehensive stress DOE test is added based on the test results. According to the impact of a single stress on the performance parameters, the key performance parameters that degrade with the continuous application of stress are screened out. Based on the degradation modeling, the appropriate stress application time is selected to ensure that the key performance parameters have sufficient degradation conditions during the DOE test implementation. This allows the impact of comprehensive stress on the key performance of lightning protection components and the coupling relationship between comprehensive stresses to be explored.

[0090] According to the coupling relationship reflected by the test data of lightning protection components, the main effects and interaction effects of each stress are quantified, and then a more reasonable comprehensive stress acceleration factor calculation method is constructed; based on the acceleration factor calculation method, the data collected in the test are transformed and processed to obtain test data suitable for existing reliability assessment methods, thereby completing the reliability assessment of lightning protection components. BRIEF DESCRIPTION OF THE DRAWINGS

[0091] Figure 1 A schematic diagram of the invention process;

[0092] Figure 2 This is a cross-sectional schematic diagram of the low-temperature step test of the invention;

[0093] Figure 3 This is a schematic cross-sectional diagram of the high-temperature step test of the invention;

[0094] Figure 4 This is a schematic cross-sectional diagram of the high temperature and high humidity test of the invention;

[0095] Figure 5 This is a cross-sectional schematic diagram of the three-axis six-degree-of-freedom random vibration step stress test of the invention;

[0096] Figure 6 Design a regression orthogonal experimental plan for the invention;

[0097] Figure 7 This is the final DOE test plan for this invention. DETAILED DESCRIPTION

[0098] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0099] In the description of the present invention, unless otherwise specified, "plurality" means two or more; terms such as "upper," "lower," "left," "right," "inner," "outer," "front end," "rear end," "head," and "tail" indicate positions or relationships based on those shown in the accompanying drawings. These terms are intended solely to facilitate the description of the present invention and simplify the description. They do not indicate or imply that the devices or components referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limiting the present invention. Furthermore, terms such as "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0100] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "connected" and "connection" should be understood in a broad sense. For example, they can refer to fixed connection, detachable connection, or integral connection; mechanical connection, electrical connection; direct connection, or indirect connection through an intermediary. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on the specific circumstances.

[0101] See also Figure 1 The reliability data mining and evaluation method of lightning protection components based on HALT test includes the following specific steps:

[0102] Step 1: Conduct a HALT test of the lightning protection component with a single step stress, as follows:

[0103] When evaluating the reliability of lightning protection components, it is first necessary to comprehensively collect the design parameters and key performance indicators of the lightning protection components. Design parameters include but are not limited to size, material, structural characteristics, etc. These parameters directly affect the key performance and tolerance of the lightning protection components. Key performance indicators are used to measure the performance of lightning protection components during operation, such as breakdown voltage, out-of-band suppression, RF S parameters, isolation, TVS surge life, etc. These indicators are crucial for evaluating the reliability of lightning protection components in actual applications.

[0104] Combining the laboratory's capabilities, a single stress step test profile was designed. A stress step test profile is a test method that systematically increases external stress (such as temperature, humidity, vibration, etc.) to observe changes in the performance of lightning protection components. By gradually increasing and changing external stress, the performance and potential failure modes of lightning protection components under different environments and usage conditions can be effectively simulated and evaluated.

[0105] When designing the test profile, the following points should be considered:

[0106] 1) Selection of stress type: According to the working principle of the lightning protection component and the expected application environment, select the appropriate stress type: temperature stress, humidity stress, vibration stress;

[0107] 2) Stress level setting: Based on the design parameters and expected application conditions of the lightning protection components, the stress levels at each stage should be reasonably set to ensure that the test is challenging but not too destructive;

[0108] 3) Stress step size: Determine the step size of the stress level increase to ensure that the test is carried out within a controllable range, avoiding excessively rapid stress increases that may cause the component to be unable to withstand, or excessively slow increases that may not fully reveal the component's performance and reliability;

[0109] 4) Monitoring and recording: During the entire test process, continuously monitor the performance changes and any abnormal conditions of the lightning protection components, and record key data for subsequent analysis and evaluation;

[0110] Compared with temperature, humidity and vibration stress, temperature cycle has less impact on lightning protection components. Therefore, only low temperature step test, high temperature step test, high temperature and high humidity test and vibration step test are considered. The test profile is as follows: Figure 2-Figure 5 ;

[0111] Step 2: Exploration of multi-stress coupling effects based on DOE experimental design;

[0112] 1) Screening key performance parameters: Preliminary screening of key performance parameters and understanding of their changing trends through single step stress HALT testing; Collecting performance data on lightning protection components under different single stresses, screening out key performance parameters with degradation trends (breakdown voltage, TVS surge life), which can, to a certain extent, reflect the response characteristics of lightning protection components under long-term stress, and establish a degradation model;

[0113] Common performance degradation linear degradation models include:

[0114] ;

[0115] ;

[0116] ;

[0117] ;

[0118] ;

[0119] 2) Design a DOE test scheme based on performance degradation. Based on the results of the single step stress HALT test, that is, the degradation and reaction rate of the key performance parameters (breakdown voltage, TVS surge life) of the lightning protection component under a single stress, determine the stress levels and the number of levels for each single stress in the DOE test scheme. On this basis, reasonably select DOE test schemes such as orthogonal design, uniform design or single regression orthogonal test to obtain key data information while reducing the number of tests and controlling test costs, which is more in line with economic requirements.

[0120] Problem description: Taking the linear regression orthogonal design as an example, for the HALT test of lightning protection components, the temperature stress A, vibration stress B, and humidity stress C are considered to have two levels for each stress, and there are interaction levels between the stresses. Select the orthogonal table L8 (2 7 ), and encode the scheme, that is, change the original 2 to -1 to obtain an orthogonal regression design scheme, such as Figure 6 As shown;

[0121] Different from the traditional DOE test scheme, this DOE test scheme needs to be based on the results of the single step stress HALT test to determine the test time at each level of each stress, to ensure that at each stress level, the key performance degradation data of the lightning protection component (breakdown voltage, TVS surge life) has at least 3% to 5% degradation. Taking formula (1) as an example, the test time at each stress level is determined. The test time of different levels of the same stress is kept consistent, and the time when the minimum level reaches the minimum degradation standard is used as the standard.

[0122] ;

[0123] The test time selection standard is to ensure that the lower limit of the level within the acceptable test time, the key performance parameters of the lightning protection component (breakdown voltage, TVS surge life) drop by 3%. Excessive vibration stress will lead to premature failure of solder joints and connections, and there is stress coupling in the comprehensive test. Therefore, when designing the DOE test, the vibration level selection needs to be conservative, such as Figure 7 As shown;

[0124] 3) Finally, collect the DOE test results, perform variance analysis and partial regression coefficient test, conduct the test according to the DOE test design, collect the degradation of the key performance parameters of the lightning protection components (breakdown voltage, TVS surge life), and use the variance analysis method to obtain the relationship between stress and key performance parameters (breakdown voltage, TVS surge life), that is, the main effect and interaction effect. Select the appropriate significance level and unify the variance analysis results to lay the foundation for the calculation of the acceleration factor;

[0125] Assume that each factor has p levels and each level has r experiments. The steps of variance analysis are as follows:

[0126] Solving for the total sum of squares of the variation :

[0127] ;

[0128] in The test results of each group; Represents the average of all test results;

[0129] Solving for the sum of squares of between-group variation :

[0130] ;

[0131] in is the average value of the test results at level i;

[0132] Solving for the within-group sum of squares of variation :

[0133] ;

[0134] Find the F test:

[0135] ;

[0136] in represents the sum of squares of the between-group variation The degree of freedom is p-1; represents the sum of squares of within-group variation The degree of freedom is p(r-1);

[0137] Look up the table to get F, compare the results, if , it is significant, otherwise it is not significant; thus, we can compare the influence of various factors on key performance parameters;

[0138] Solve the multiple linear regression equation for the test results:

[0139] ;

[0140] Standardize the coefficients of the multiple linear regression equation:

[0141] ;

[0142] in, is the standardized regression coefficient; is the independent variable The standard deviation of is the standard deviation of the dependent variable y;

[0143] Step 3: Calculation of acceleration factors for single stress and combined stress;

[0144] 1) Temperature stress acceleration factor, using the Arrehenius model as the temperature stress acceleration model, the temperature acceleration stress T s Acceleration factor A under relative normal stress T0 T The calculation formula is:

[0145] ;

[0146] Where, It is the product life characteristics under normal temperature stress; is the product life characteristics under accelerated stress; E a is the activation energy, in eV; k is the Boltzmann constant, k=8.617x×10 -5 eV / K; T0 is the absolute temperature value of normal temperature stress, unit is K; T s is the absolute temperature value of the accelerated temperature stress, in K;

[0147] 2) Humidity acceleration factor. Using the Peck model as the acceleration model, the calculation formula for the humidity stress acceleration factor is:

[0148] ;

[0149] RH0 is normal humidity stress; RH S To accelerate humidity stress; n is 2~3;

[0150] 3) Vibration acceleration factor, using the inverse power law model as the acceleration model, the vibration stress acceleration factor calculation method is:

[0151] ;

[0152] W0 is the normal vibration stress level; W1 is the accelerated vibration stress level; V0 is the test time under the normal vibration stress level; V1 is the test time under the accelerated vibration stress level;

[0153] 4) Acceleration factor calculation method based on temperature stress. Under comprehensive environmental stress, the effects of various stresses on products have a coupling mechanism, which can be divided into mutual independence, mutual inhibition, and mutual promotion. In the independent case, the acceleration factor can be the multiplication of the acceleration factors of a single stress. However, in the case of inhibition or promotion, the magnitude of the mutual promotion and inhibition effects needs to be quantified.

[0154] The Arrehenius model has been widely used in accelerated reliability assessments of products. Since lightning protection components are most affected by temperature stress, a single temperature stress acceleration factor is used as the benchmark. Combined with the normalized regression coefficient from step 3), the acceleration factor calculation formula is:

[0155] ;

[0156] The acceleration factors under other single accelerated stress and comprehensive accelerated stress can be obtained from the above formula: ;

[0157] Step 4: Data conversion and reliability assessment based on acceleration factors;

[0158] Select the test data corresponding to the appropriate cut-off stress level to ensure that the lightning protection components meet the basic assumptions of the accelerated test under various stresses, that is, the failure mechanism is consistent. Based on the selection of the cut-off stress level, the problem of test data being unusable due to excessive stress and inconsistent product failure mechanism is avoided.

[0159] The test time before the cutoff stress level is converted. Tests under different stress levels are equivalent to the test time under the same stress level. Generally, the stress lower limit of the HALT test and DOE test is selected. The test time conversion formula for the HALT test is as follows:

[0160] ;

[0161] in represents the actual test time under the pth stress step, with a total of M steps; represents the acceleration factor of the first stress step based on the temperature accelerated stress; represents the acceleration factor of the p-th step based on temperature accelerated stress; The test time is converted based on the first step acceleration factor;

[0162] After final conversion, the accelerated test data under a certain stress value (the lower limit of stress in the DOE test design) will be obtained. The reliability evaluation results of the product can be obtained by following the steps below:

[0163] 1) Key performance data degradation modeling: fitting the function of degradation data and time to obtain the key performance parameter degradation model;

[0164] 2) Set the failure threshold and calculate the pseudo-failure life value of the lightning protection component based on the degradation model;

[0165] 3) Conduct hypothesis testing on the pseudo-failure life value of the sample to determine whether it satisfies normal distribution, Wei_bull distribution, lognormal distribution, etc.;

[0166] 4) Calculate the reliability and characteristic life according to the corresponding distribution function, taking the normal distribution as an example;

[0167] 5) Based on the acceleration factor, convert the characteristic life value of the product under normal stress.

[0168] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. A method for reliability data mining and evaluation of lightning protection components based on HALT test, characterized by: The specific steps include: S1. Single step stress HALT test of lightning protection components is as follows: When conducting a reliability assessment of a lightning protection component, it is first necessary to comprehensively collect the design parameters and key performance indicators of the lightning protection component. Design parameters include but are not limited to dimensions, materials, and structural characteristics. Key performance indicators are used to measure the performance of the lightning protection component during operation. Based on the laboratory's capabilities, a single stress step test profile is designed. S2: Investigation of multi-stress coupling effects based on DOE experimental design, as follows: S201. Screen key performance parameters. Preliminarily screen key performance parameters and understand the changing trends of these performance parameters through a single step stress HALT test. Collect performance data on lightning protection components under different single stresses, screen out key performance parameters with degradation trends, thereby reflecting the response characteristics of lightning protection components under long-term stress, and establish a performance degradation model. S202. Design a DOE test plan based on performance degradation. Based on the degradation and reaction rate of the key performance parameters of the lightning protection component under a single stress, determine the stress levels and number of levels for each single stress in the DOE test plan. On this basis, reasonably select a DOE test plan of orthogonal design, uniform design, or single regression orthogonal test. This DOE test plan needs to be based on the results of the single step stress HALT test to determine the test time for each level of each stress, to ensure that at each stress level, the degradation data of the key performance of the lightning protection component has at least 3% to 5% degradation, and to determine the test time for each stress level. The test time for different levels of each stress should be consistent, and the time when the minimum level reaches the minimum degradation standard shall prevail. ； in, is the performance indicator, For initial performance, is the degradation rate, For time; S203. Collect DOE test results, perform variance analysis and partial regression coefficient test, conduct tests according to the DOE test design, collect the degradation of key performance parameters of lightning protection components, use variance analysis method to obtain the relationship between stress and key performance parameters, namely the main effect and interaction effect, select the significance level, unify the variance analysis results, and lay the foundation for the calculation of acceleration factors; S3. Calculation of acceleration factors for single stress and combined stress is as follows: S301, temperature stress acceleration factor, using the Arrhenius model as the temperature stress acceleration model, temperature accelerated stress T s Acceleration factor A under relative normal stress T0 T The calculation formula is: ； Where, It is the product life characteristics under normal temperature stress; is the product life characteristics under accelerated stress; E a is the activation energy, in eV; k is the Boltzmann constant, k=8.617x×10 -5 eV / K; T0 is the absolute temperature value of normal temperature stress, unit is K; T s is the absolute temperature value of the accelerated temperature stress, in K; S302, humidity acceleration factor, using the Peck model as the acceleration model, the humidity stress acceleration factor calculation formula is: ； RH0 is normal humidity stress; RH S To accelerate humidity stress; n is 2~3; S303, vibration acceleration factor, using the inverse power law model as the acceleration model, the vibration stress acceleration factor calculation method is: ； W0 is the normal vibration stress level; W1 is the accelerated vibration stress level; V0 is the test time under the normal vibration stress level; V1 is the test time under the accelerated vibration stress level; S304. Calculation method for acceleration factors based on temperature stress. Under comprehensive environmental stress, the effects of various stresses on products are coupled, classified as mutual independence, mutual inhibition, and mutual promotion. In the independent case, the acceleration factor is the multiplication of the acceleration factors of the individual stresses. However, in the inhibition / promotion case, the magnitude of the mutual promotion and inhibition effects needs to be quantified. A single temperature stress acceleration factor is selected as the benchmark, and the acceleration factor calculation formula is obtained by combining the normalized regression coefficient in step 2: ； The acceleration factors under other single accelerated stress and comprehensive accelerated stress are obtained from the above formula: ; S4. Data conversion and reliability assessment based on acceleration factors are as follows: Select the test data corresponding to the cut-off stress level to ensure that the lightning protection components meet the basic assumptions of the accelerated test under various stresses, that is, the failure mechanism is consistent; The test time before the cutoff stress level is converted. Tests under different stress levels are equivalent to the test time under the same stress level. The stress lower limit value of the HALT test and DOE test is selected. The test time conversion formula for the HALT test is as follows: ； in, represents the actual test time under the pth stress step, with a total of M steps; represents the acceleration factor of the first stress step based on the temperature accelerated stress; represents the acceleration factor of the p-th step based on temperature accelerated stress; The test time is converted based on the first step acceleration factor; After final conversion, the accelerated test data under a certain stress value will be obtained, that is, the reliability evaluation result of the product will be obtained.

2. The method for reliability data mining and evaluation of lightning protection components based on HALT test according to claim 1 is characterized in that: The stress step test profile in step S1 is specifically a test method for systematically increasing external stress and observing changes in the performance of lightning protection components. By gradually increasing and changing the external stress, the performance and potential failure modes of the lightning protection components under different environments and usage conditions are effectively simulated and evaluated.

3. The method for reliability data mining and evaluation of lightning protection components based on HALT test according to claim 2, characterized in that: The stress step test profile specifically includes a low temperature step test, a high temperature step test, a high temperature and high humidity test, and a vibration step test; When designing the test profile, the following points should be considered: S101. Selection of stress type: According to the working principle of the lightning protection component and the expected application environment, select the stress type: temperature stress, humidity stress, vibration stress; S102, Stress level setting: Based on the design parameters and expected application conditions of the lightning protection components, the stress levels at each stage should be reasonably set to ensure that the test is challenging but not too destructive; S103. Stress step size: Determine the step size of the stress level increase to ensure that the test is carried out within a controllable range, avoiding excessively rapid stress increases that may cause the component to be unable to withstand, or excessively slow increases that may not fully reveal the performance and reliability of the component; S104. Monitoring and recording: During the entire test process, continuously monitor the performance changes and any abnormal conditions of the lightning protection components, and record key data for subsequent analysis and evaluation.

4. The method for reliability data mining and evaluation of lightning protection components based on HALT test according to claim 1 is characterized in that: The performance degradation linear degradation model in step S202 is as follows: ； in, is the performance indicator, For initial performance, is the degradation rate, if <0, indicating that performance changes over time Linear decrease; if >0, the performance improves; ； in, is the logarithmic transformation of time; ； in, is the logarithmic transformation of the performance, is the power law index, if <0, indicating that performance changes over time Power law decay; ； in, For exponential transformation, performance changes exponentially with time t; ； in, is a scale parameter used to control the degradation rate.

5. The method for reliability data mining and evaluation of lightning protection components based on HALT test according to claim 1 is characterized in that: The variance analysis method in step S203 is specifically as follows: Assume that each factor has p levels and each level has r experiments. The steps of variance analysis are as follows: Solving for the total sum of squares of the variation : ； in The test results of each group; Represents the average of all test results; Solving for the sum of squares of between-group variation : ； in is the average value of the test results at level i; Solving for the within-group sum of squares of variation : ； in is the average value of the test results at level i; Find the F test: ； in represents the sum of squares of the between-group variation The degree of freedom is p-1; represents the sum of squares of within-group variation The degree of freedom is p(r-1); Look up the table to get F, compare the results, if , it is significant, otherwise it is not significant; thus, we can compare the influence of various factors on key performance parameters; Solve the multiple linear regression equation for the test results: ； Standardize the coefficients of the multiple linear regression equation: ； in, is the standardized regression coefficient; is the independent variable The standard deviation of is the standard deviation of the dependent variable y.

6. The method for reliability data mining and evaluation of lightning protection components based on HALT test according to claim 1 is characterized in that: The process of obtaining the reliability evaluation result in step S4 is as follows: S401, key performance data degradation modeling, fitting the function of degradation data and time to obtain a key performance parameter degradation model; S402: Setting a failure threshold and calculating a pseudo-failure life value of the lightning protection component according to a degradation model; S403, performing a hypothesis test on the pseudo failure life value of the sample to determine whether it satisfies normal distribution, Wei_bull distribution, or lognormal distribution; S404. Calculate reliability and characteristic life according to the corresponding distribution function, taking normal distribution as an example; S405. Based on the acceleration factor, calculate the characteristic life value of the product under normal stress.

Images