Reliability enhancement test relative reliability quantification assessment method, system and apparatus

Abstract

The application discloses a kind of reliability enhancement test relative reliability quantitative evaluation method, system and equipment, the method includes based on interference theory, interference model is constructed to product environmental stress and reliability enhancement test tolerance limit stress;Product reliability enhancement test tolerance limit stress and environmental stress are respectively subjected to distribution hypothesis test, study the distribution type that it obeys;The distribution function of product reliability enhancement test tolerance limit stress and environmental stress is substituted into the interference model, the reliability of product is calculated;According to the reliability of new product and old product calculated, the relative reliability of new product relative to old product in reliability enhancement test is obtained.The application establishes relative reliability evaluation method by analyzing the tolerance limit stress distribution in the process of product reliability enhancement test, solves the applicability problem of traditional reliability evaluation method in the process of reliability enhancement test evaluation.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of reliability enhancement test, and particularly relates to a relative reliability quantification evaluation method, system and device for reliability enhancement test. BACKGROUND

[0002] Reliability enhancement test refers to sequentially performing low-temperature step stress test, high-temperature step stress test, rapid temperature change cycle stress test, vibration step stress test and comprehensive environmental stress test on a test sample, and gradually increasing the test stress applied to the test sample to quickly expose potential design defects and determine the stress limit that the product can withstand. In recent years, reliability enhancement test has been widely used in the reliability test of electronic equipment, especially in the fields of communication, energy, aviation and aerospace. Internationally, reliability enhancement test technology has been relatively mature, and some standards on how to carry out reliability enhancement test have been formed. In China, the research on reliability enhancement test technology started relatively late, and reliability enhancement test is in the state of following foreign countries and preliminary application in typical products.

[0003] Reliability enhancement test can quickly stimulate product defects in a short time, improve the tolerance of products in the use environment, enhance the working limit and breaking limit of products, and shorten the development time of products. However, because the reliability enhancement test is short in time, the degradation test data obtained is limited, it is difficult to directly obtain the reliability index of the product by relying on a large amount of test data for analysis as in traditional reliability test. One of the advantages of reliability enhancement test is to shorten the product development cycle. At the beginning of product design or during testing, a large number of samples are not usually invested for testing, and small sample is also the difficulty of reliability evaluation. So far, there is no mature reliability quantification evaluation technology for reliability enhancement test to directly obtain the reliability index of the product at home and abroad.

[0004] Therefore, in view of the problems of less reliability enhancement test data and inability to directly quantification evaluation, it is necessary to find an indirect evaluation technology for quantification evaluation of product reliability. This evaluation technology can indirectly evaluate the reliability index of the product without relying on a large number of samples and a large amount of test data.

[0005] At present, there are few studies on reliability quantification evaluation technology for reliability enhancement test, and more studies on test methods. There is no mature reliability quantification evaluation technology, and no reliability index of reliability enhancement test is given. There are more studies on reliability quantification evaluation technology for products. The mainstream direction is the reliability quantification evaluation research of accelerated life test and accelerated degradation test, and there are also some small amount of research, for example, using stress-strength interference model to quantitatively analyze reliability enhancement test, or directly applying accelerated test quantification evaluation method to reliability enhancement test evaluation. However, these technologies have the following disadvantages:

[0006] First, different products have different stress tolerance and different distribution types of the stress to which the products are subjected, which leads to unreasonable selection of stress distribution types in the evaluation method and low accuracy of the evaluation results.

[0007] Second, the evaluation direction is not appropriate, and the reliability quantification evaluation of the reliability enhancement test is directly performed in the same way as the acceleration test, but the mechanism of the reliability enhancement test is not consistent with that of the acceleration test, and the data of the reliability enhancement test are less, so the quantification evaluation method of the acceleration test cannot be directly applied to the evaluation of the reliability enhancement test to judge the accuracy of the reliability index. SUMMARY

[0008] In order to solve the problem of low accuracy of the reliability quantification evaluation results of the existing product reliability enhancement test, the present application provides a reliability enhancement test relative reliability quantification evaluation method, system and device, which fully utilizes the product reliability enhancement test tolerance limit stress and the environmental stress to which the product is subjected, establishes a reliability quantification evaluation technology based on the interference theory, and considers the universal applicability of the product tolerance limit stress in multiple distribution types, so as to realize accurate evaluation of the relative reliability of the reliability enhancement test on the basis of limited test data.

[0009] The present application is realized by the following technical solutions:

[0010] A reliability enhancement test relative reliability quantification evaluation method, the method comprising:

[0011] Based on the interference theory, an interference model of the product environmental stress and the reliability enhancement test tolerance limit stress is constructed;

[0012] The distribution hypothesis test is performed on the product reliability enhancement test tolerance limit stress and the environmental stress, and the distribution types to which they are subjected are researched;

[0013] The distribution functions of the product reliability enhancement test tolerance limit stress and the environmental stress are substituted into the interference model, and the reliability of the product is calculated;

[0014] According to the calculated reliability of the new product and the old product, the relative reliability of the new product relative to the old product in the reliability enhancement test is obtained.

[0015] In some embodiments, the interference model constructed by the method is that the probability density curve of the product reliability enhancement test tolerance limit stress and the probability density curve of the environmental stress cross and overlap, and the interference area is not normal working of the product.

[0016] In some embodiments, the method uses K-S test to perform distribution hypothesis test on the product reliability enhancement test limit stress and / or environmental stress, and determines the distribution type to which the limit stress and / or environmental stress can be subjected.

[0017] After determining the distribution type, parameter estimation of the stress population is performed, wherein the parameter estimation method is determined according to the distribution type.

[0018] In some embodiments, when the product reliability enhancement test limit stress and environmental stress are subjected to normal distribution, point estimation of the mean value is performed by using maximum likelihood method, and point estimation of the mean square deviation is performed by using minimum variance method.

[0019] In some embodiments, the reliability of the product is the probability that the product works normally in the entire environmental stress range.

[0020] In some embodiments, the reliability calculation formula of the product is:

[0021]

[0022] In the formula, R is the reliability of the product, P(S>E) is the probability that the product reliability enhancement test limit stress is greater than the environmental stress, h(E) is the probability density of the environmental stress, and f(S) is the probability density of the product reliability enhancement test limit stress.

[0023] In some embodiments, the relative reliability is the ratio of the reliability of the new product to the reliability of the old product.

[0024] In a second aspect, the present application provides a reliability enhancement test relative reliability quantitative evaluation system, which comprises:

[0025] A model construction module, which constructs an interference model of the product environmental stress and the reliability enhancement test limit stress based on interference theory;

[0026] A test module, which performs distribution hypothesis test on the product reliability enhancement test limit stress and environmental stress, respectively, and studies the distribution type to which the limit stress and / or environmental stress is subjected.

[0027] A calculation module, which substitutes the distribution function of the product reliability enhancement test limit stress and environmental stress into the interference model, and calculates the reliability of the product.

[0028] A quantitative module, which obtains the reliability enhancement test relative reliability of the new product with respect to the old product according to the calculated reliability of the new product and the old product.

[0029] In a third aspect, the present application provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method of the present application when executing the computer program.

[0030] In a fourth aspect, the present application provides a computer readable storage medium, which stores a computer program, and the computer program implements the steps of the method of the present application when executed by a processor.

[0031] The relative reliability quantitative evaluation method, system and device for reliability enhancement test provided by the present application can realize fast and accurate evaluation of the relative reliability of the product reliability enhancement test, and solve the applicability problem of the traditional reliability evaluation method in the reliability enhancement test evaluation process.

[0032] The relative reliability quantitative evaluation method, system and device for reliability enhancement test provided by the present application adopts the K-S test method for distribution hypothesis testing, which is suitable for small sample occasions, and therefore is particularly suitable for application occasions with small sample size in reliability enhancement test. BRIEF DESCRIPTION OF DRAWINGS

[0033] The accompanying drawings, which are included to provide a further understanding of the embodiments of the present application and constitute a part of the application, illustrate embodiments of the present application and are used to explain the principle of the present application, but do not limit the present application. In the drawings:

[0034] Figure 1 The method flowchart of the embodiments of the present application is shown in the figure;

[0035] Figure 2 The schematic diagram of the change process of the product tolerance limit stress subjected to the reliability enhancement test is shown in the figure;

[0036] Figure 3 The interference model of the environmental stress and the tolerance limit stress of the embodiments of the present application is shown in the figure;

[0037] Figure 4 The system principle block diagram of the embodiments of the present application is shown in the figure. DETAILED DESCRIPTION

[0038] In order to make the purpose, technical scheme and advantages of the present application clearer, the present application is further described in detail below with reference to the embodiments and the accompanying drawings, and the illustrative embodiments of the present application and the description thereof are used to explain the present application, but not to limit the present application.

[0039] Embodiment:

[0040] In view of the problem of low accuracy of reliability quantitative evaluation of product reliability enhancement test, the embodiment provides a relative reliability quantitative evaluation method of reliability enhancement test. The method provided in the embodiment is based on the interference theory, establishes an interference model of product environmental stress and tolerance baseline stress, and adopts hypothesis testing to make the product tolerance limit stress distribution more universal and applicable, so that the reliability quantitative evaluation of reliability enhancement test is realized quickly and accurately.

[0041] As shown in Figure 1 , the method provided in the embodiment specifically comprises the following steps:

[0042] Step 1, based on the interference theory, an interference model of product environmental stress and reliability enhancement test tolerance limit stress is constructed;

[0043] Step 2, distribution hypothesis testing is performed on the product reliability enhancement test tolerance limit stress and environmental stress respectively, and the distribution type obeyed thereby is researched;

[0044] Step 3, the distribution functions of the product reliability enhancement test tolerance limit stress and environmental stress are substituted into the interference model, and the reliability of the product is calculated;

[0045] Step 4, the reliability of the new product and the old product is obtained through the above step, and the relative reliability of the new product to the old product in the reliability enhancement test is obtained.

[0046] A group of products are subjected to reliability enhancement test to obtain a new product meeting a certain design target, and the new product is subjected to reliability enhancement test again to verify whether it meets the design target. The tolerance limit stress and the environmental stress of the product during the reliability enhancement test are collected. It is assumed that the tolerance limit stress collected from n samples is S i , i = 1, 2, …, n; and the environmental stress of the product is E i , i = 1, 2, …, n.

[0047] Further, the interference model construction process is specifically as follows:

[0048] After the product is subjected to the reliability enhancement test, the tolerance limit stress is improved, and the new product is obtained, and the change process is as shown in Figure 2 . The probability density curve f(S) of the old product at a certain time is as shown on the left side of Figure 2 . After the reliability enhancement test, the new product is obtained, the tolerance limit stress of the new product is increased, and the probability density curves of the new and old products at the same time are as shown on the right side of Figure 2 . Among them, the probability density curve of the old product is as shown in the lower right part of Figure 2 , and the probability density curve g(δ) of the new product is as shown in Figure 2The upper right part is shown.

[0049] The condition for the product to work normally is that the self-tolerance limit stress (set as S) is greater than the environmental stress (set as E) in which the product is located. However, even if the average of the product tolerance limit stress S is greater than the average of the environmental stress E, the situation that the individual product tolerance limit stress S is less than the environmental stress E may occur, that is, the interference superposition occurs, and the situation that the product cannot work normally occurs, even if the new product is obtained by enhancing the old product after the reliability enhancement test. Therefore, the product interference region cannot work normally, that is, the probability density curve of the product tolerance limit stress and the probability density curve of the environmental stress cross and overlap, and the interference occurs, and therefore the interference model of the environmental stress and the tolerance limit stress obtained by construction is as shown in Figure 3 , wherein h(E) is the probability density curve of the environmental stress, and f(S) is the probability density curve of the tolerance limit stress.

[0050] Further, the reliability enhancement test tolerance limit stress level of the same type of product population should be basically consistent. Due to the random fluctuation between products, the reliability enhancement test tolerance limit stress is not the same, but the randomness can be described by a certain specific distribution. Similarly, the environmental stress in which the product is located is also the same. Therefore, the present embodiment performs distribution hypothesis testing on the tolerance limit stress and the environmental stress, that is, when the tolerance limit stress distribution needs to be tested, the environmental stress needs to be fixed or the tolerance limit stress distribution of different product individuals is tested under a specific environmental stress level, and similarly, when the environmental stress distribution needs to be tested, the tolerance limit stress needs to be fixed or the environmental stress distribution of different product individuals is tested under a specific tolerance limit stress level. Further, the environmental stress distribution type is usually a normal distribution, so only the distribution parameters need to be estimated; if the environmental stress distribution does not satisfy the normal distribution, the distribution type hypothesis test needs to be performed according to the environmental stress data sample. Further, due to the small sample size of the reliability enhancement test, the present embodiment uses the K-S test method or other small sample test methods to perform distribution hypothesis testing on the tolerance limit stress and / or the environmental stress, and determines the distribution type that can be subjected to.

[0051] The K-S test method is a short name of the Kolmogorov-Smirnov test method, also known as the D test method, which is used to test the fitting degree of the assumed distribution of the collected data and the actual distribution. When the test statistic D n ≤d n,αWherein, a is the confidence level, the hypothesis distribution is consistent with the true probability distribution of the population in the statistical sense; otherwise, the hypothesis distribution is rejected. After using K-S test method to obtain the distribution type of the sample, the parameter estimation of the stress population is carried out, and the parameter estimation method is different for different sample distribution types. Generally, moment estimation, least square estimation, maximum likelihood estimation and Bayesian estimation can be used.

[0052] For example, when the environmental stress and the tolerance limit stress data (here, the related data is denoted as: x i , i = 1, 2, …, n) obeys normal distribution, the point estimation of mean value is Using maximum likelihood method, the point estimation of mean square deviation is Using minimum variance unbiased estimation, the specific calculation formula is as follows:

[0053]

[0054]

[0055] Further, according to the distribution type of the product reliability enhancement test tolerance limit stress and environmental stress, the distribution function is substituted into the interference model of environmental stress and tolerance limit stress, and the reliability of the product based on the interference theory is obtained. The probability of the product working normally in the whole environmental stress interval is the reliability R of the product, and the specific calculation formula is as follows:

[0056]

[0057] In the formula, R is the reliability of the product, P (S > E) is the probability that the product reliability enhancement test tolerance limit stress is greater than the environmental stress, h (E) is the probability density of the environmental stress, and f (S) is the probability density of the product reliability enhancement tolerance limit stress.

[0058] Further, through step 3, the reliability of the new product and the old product can be obtained, and the ratio of the reliability of the new product and the old product is the relative reliability of the reliability enhancement test based on the interference theory. Wherein, the reliability enhancement test tolerance limit stress of the old product is set as S1, and the reliability enhancement test tolerance limit stress of the new product is set as S2, then the relative reliability R' of the new product to the old product based on the interference theory is calculated according to the following formula:

[0059]

[0060] The method provided in the embodiment is based on the interference theory, and an interference model of the environmental stress of the product and the endurance limit stress in the reliability enhancement test is established.

[0061] The method provided in the embodiment adopts the K-S test method to perform the distribution hypothesis test, so that the reliability evaluation technology is more suitable for small sample occasions, and thus is suitable for the reliability enhancement test with a small sample quantity.

[0062] Based on the same technical concept, the embodiment further provides a reliability enhancement test relative reliability quantitative evaluation system, as shown in Figure 4 The system provided in the embodiment specifically comprises:

[0063] A model construction module, which is based on the interference theory and constructs an interference model of the environmental stress of the product and the endurance limit stress in the reliability enhancement test.

[0064] A test module, which performs the distribution hypothesis test on the endurance limit stress and the environmental stress distribution of the product in the reliability enhancement test, and studies the distribution type thereof.

[0065] A calculation module, which substitutes the distribution functions of the endurance limit stress and the environmental stress of the product in the reliability enhancement test into the interference model to obtain the reliability of the product.

[0066] A quantitative module, which obtains the reliability enhancement test relative reliability of the new product relative to the old product according to the respective reliabilities of the new product and the old product obtained by the calculation module.

[0067] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program codes.

[0068] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 one or more flow or blocks

[0069] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 one or more flow or blocks

[0070] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 one or more flow or blocks

[0071] The above description is only specific implementation of the present application, and is not intended to limit the protection scope of the present application, and any modification, equivalent replacement, improvement, etc. within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for quantitatively evaluating relative reliability of reliability enhancement test, characterized by, The method comprises: collecting the limit stress and environmental stress of a product during a reliability enhancement test; constructing an interference model of the limit stress and environmental stress of the product based on interference theory; performing distribution hypothesis testing on the limit stress and environmental stress of the product to study the distribution type thereof; substituting the distribution function of the limit stress and environmental stress of the product into the interference model to calculate the reliability of the product; obtaining the relative reliability of the new product relative to the old product according to the calculated reliability of the new product and the old product; and the interference model is constructed as follows: the probability density curve of the limit stress of the product and the probability density curve of the environmental stress of the product cross and overlap, and the product cannot work normally in the interference area; and the probability density curve of the limit stress of the product changes with time. ； wherein, is the relative reliability, is the new product reliability-strengthened test limit stress, is the old product reliability-strengthened test limit stress, is the probability density of the environmental stress, is the probability density of the old product reliability-strengthened limit stress, is the probability density of the new product reliability-strengthened limit stress, is the environmental stress.

2. The method of claim 1, wherein the method comprises: The relative reliability calculation formula of the new product relative to the old product is as follows: The method uses K-S testing method to perform distribution hypothesis testing on the limit stress and / or environmental stress of the product to determine the distribution type thereof.

3. The method of claim 2, wherein the method comprises: determining the relative reliability of the reliability-strengthened test based on the number of the reliability-strengthened test and the number of the reliability-strengthened test that passed the reliability-strengthened test. After the distribution type is determined, parameter estimation of the stress population is performed, wherein the parameter estimation method is determined according to the distribution type.

4. The method of claim 1-3, wherein, When the limit stress and environmental stress of the product during the reliability enhancement test obey normal distribution, the point estimation of the mean value is performed by maximum likelihood method, and the point estimation of the mean square deviation is performed by minimum variance method.

5. The method of claim 1-3, wherein, The reliability of the product is the probability that the product works normally in the entire environmental stress range. ； wherein is the reliability of the product, is the probability that the limit stress of the product reliability enhancement test is greater than the environmental stress, is the probability density of the environmental stress, is the probability density of the limit stress of the product reliability enhancement.

6. A system for quantitatively evaluating relative reliability of reliability enhancement tests, characterized by, The reliability calculation formula of the product is as follows: The system comprises: a model construction module, which is configured to collect the limit stress and environmental stress of a product during a reliability enhancement test, and construct an interference model of the limit stress and environmental stress of the product based on interference theory; a testing module, which is configured to perform distribution hypothesis testing on the limit stress and environmental stress of the product to study the distribution type thereof; a calculation module, which is configured to substitute the distribution function of the limit stress and environmental stress of the product into the interference model to calculate the reliability of the product; a quantification module, which is configured to obtain the relative reliability of the new product relative to the old product according to the calculated reliability of the new product and the old product; and the interference model is constructed as follows: the probability density curve of the limit stress of the product and the probability density curve of the environmental stress of the product cross and overlap, and the product cannot work normally in the interference area. ； wherein, is the relative reliability, is the new product reliability-strengthened test limit stress, is the old product reliability-strengthened test limit stress, is the probability density of the environmental stress, is the probability density of the old product reliability-strengthened limit stress, is the probability density of the new product reliability-strengthened limit stress, is the environmental stress. 7.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-6. The relative reliability calculation formula of the new product relative to the old product is as follows:

8. A computer readable storage medium having stored thereon a computer program, characterized in that, The processor implements the steps of the method according to any one of claims 1-5 when executing the computer program. The computer program implements the steps of the method according to any one of claims 1-5 when executed by the processor.

Images