Comprehensive evaluation method for storage life of complex mechanical and electrical product based on life probability weighting

Abstract

The invention belongs to the technical field of mechanical and electrical products, and discloses a comprehensive evaluation method for the storage life of a complex mechanical and electrical product based on life probability weighting, which comprises the following steps: acquiring a newly produced product, a product weak link and iterative product index data, and operating a stepping temperature stress accelerated life test according to a selected stress level value; obtaining life data of each product to be evaluated; according to the service life data type of the to-be-evaluated product, combining an Arrhenius acceleration model and Weibull distribution, estimating model parameters by adopting maximum likelihood estimation and a least square method to obtain estimated values of the shape parameters and the characteristic service life of the product under normal stress, and respectively evaluating the storage service life of the product; and establishing a complex mechanical and electrical product life probability weighting model to comprehensively evaluate the storage life of the complex mechanical and electrical product. And the reliability and storage life evaluation precision of complex electromechanical products are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of electromechanical products, and in particular relates to a comprehensive evaluation method for storage life of complex electromechanical products based on life probability weighting. Background Art

[0002] With the innovation and development of modern manufacturing processes and technologies, the level of independent innovation, independent research and development, and independent control of high-reliability and long-life products is getting higher and higher, which requires the reliability and storage life assessment accuracy of complex electromechanical products with high reliability and long life to be higher and higher. Complex electromechanical products with high reliability and long life are generally stored for a long time and used once; most of the time is in the storage stage, and a small part of the time is in the transportation and handling turnover period, and the action time is only a few seconds, such as ammunition, artillery shells, mortars, missiles, etc. Therefore, it is of great significance to study the storage life of such products.

[0003] Traditional storage life assessment methods for complex electromechanical products only consider a single type of product. Traditional reliability modeling and storage life assessment methods usually only consider newly produced products, conduct accelerated life tests on newly produced products, obtain life data, and then evaluate product life. However, traditional storage life assessment methods ignore the impact of iterative products and product weaknesses on the storage life assessment of complex electromechanical products, resulting in difficulty in ensuring the accuracy of storage life assessment of complex electromechanical products. In addition, for high-reliability, long-life products that have not yet reached their lifespan and have been newly finalized, their actual storage life is difficult to obtain because they have not reached the storage period specified in the indicators. At this time, traditional storage life assessment methods are difficult to obtain effective assessment results.

[0004] As the reliability requirements of complex electromechanical products become higher and higher, comprehensively evaluating the reliability and storage life of complex electromechanical products by considering the storage life of new products, product weaknesses, and iterative products provides a new idea for improving product reliability and evaluation accuracy. Therefore, it is urgent to provide a storage life evaluation method for complex electromechanical products that comprehensively considers the reliability of new products, product weaknesses, and iterative products. Summary of the invention

[0005] In order to solve the technical problem that the traditional storage life assessment method of complex electromechanical products in the prior art ignores the impact of iterative products and product weaknesses on the storage life assessment of complex electromechanical products, the present invention provides a comprehensive storage life assessment method for complex electromechanical products based on life probability weighting, which is particularly suitable for storage life assessment of products that have not reached their lifespan, are newly finalized, have high reliability, and have long lifespans. The storage life assessment accuracy of complex electromechanical products is improved.

[0006] The present invention is achieved by adopting the following technical solutions:

[0007] Comprehensive evaluation method for storage life of complex mechatronic products based on life probability weighting, including:

[0008] A1: Obtain the index data of newly produced products, product weak links and iterative products, and run a step temperature stress accelerated life test according to the selected stress level value to obtain the life data of each product to be evaluated;

[0009] A2: According to the type of life data of the product to be evaluated, combine the Arrhenius acceleration model and the Weibull distribution to obtain the characteristic life of the product, use the maximum likelihood estimation and the least square method to estimate the model parameters, obtain the maximum likelihood estimation of the shape parameter and the characteristic life of the product under normal stress, and evaluate the storage life of the product under normal stress level; Obtain the reliability function and storage life of newly produced products, product weak links, and iterative products;

[0010] A3: Establish a life probability weighting model for complex mechatronic products, calculate the product probability competition weight, and comprehensively evaluate the storage life of complex mechatronic products.

[0011] Further, the operation of the step temperature stress accelerated life test according to the selected stress level value includes: first select a group of accelerated stress levels, place a set number of samples at the first stress level at the beginning of the test for accelerated testing; and detect the product performance parameters; after the accelerated time, increase the stress level to the second stress level, place the unfailed samples at the second stress level to continue the life test, and take out the samples for performance detection during the continued test; increase the stress level in turn according to the step stress accelerated life test, and stop the test when the specified censoring time of the test is reached.

[0012] Further, the combination of the Arrhenius acceleration model formula is as follows:

[0013]

[0014] Among them, η i , i = 1, 2,..., l is the characteristic life of the product, l is the number of stress levels; a, b are estimated parameters; Ti is the absolute temperature.

[0015] Further, combined with the Weibull distribution, the formula for the failure distribution function after time conversion at each set stress level is as follows:

[0016]

[0017] Among them, m i is the shape parameter, η i is the scale parameter under normal stress level.

[0018] Furthermore, the maximum likelihood estimation method is used to calculate the partial derivatives of the shape parameter and the scale parameter of the log-likelihood function to obtain the estimated values ​​of the shape parameter and the scale parameter; and then the estimated values ​​of the parameters a and b are obtained by the least squares method based on the Arrhenius acceleration model formula and the estimated value of the scale parameter.

[0019] Furthermore, the maximum likelihood estimation of the shape parameters and the characteristic life of the product under normal stress includes: obtaining the likelihood function of the life data under each stress according to the failure distribution function converted by the reference time, the number of samples and the number of failures:

[0020]

[0021] Find the logarithm of the likelihood function.

[0022] Furthermore, the estimated values ​​of parameters a and b are calculated using the least squares method, and the logarithms of both sides of the Arrhenius acceleration model formula (1) are taken:

[0023]

[0024] Let y = lnη i , Get the estimated values ​​of parameters a and b.

[0025] Furthermore, the estimated values ​​of shape parameters and scale parameters obtained according to the estimation model are used to evaluate the characteristic life under normal stress level.

[0026] Furthermore, the combination of the Arrhenius acceleration model and the Weibull distribution includes: under different stress levels, the product failure mechanism is the same, that is, under each stress, the shape parameter in the life distribution remains unchanged, that is, m 0 =m 1 =…=m l .

[0027] Furthermore, establishing the life probability weighted model of the complex electromechanical product includes: obtaining the reliability functions of newly produced products, weak links and iterative products, and then obtaining the storage life of each type of newly produced products, product weak links and iterative products at a given reliability based on the given reliability function, and using the storage life and reliability function to calculate the probability competition weight of each product to obtain a comprehensive evaluation of the storage life of the complex electromechanical product.

[0028] Furthermore, the formula for establishing a life probability weighted model for complex electromechanical products is:

[0029]

[0030] Among them, R,j For jAt the moment, the storage life of complex electromechanical products when the reliability is R; n is the number of products of each type to be weighted, ω i,j is the probability competition weight, that is, at t j The normalized value of the failure probability of the i-th type of product at time T R,i is the storage life of the i-th type product when the reliability is R.

[0031] Further, calculate the product probability competition weight ω i,j The formula is as follows:

[0032]

[0033] Among them, R i (t j ) is at t j The storage reliability of the i-th type of product at the moment is used to calculate the product probability competition weight based on the reliability function R(t).

[0034] Furthermore, the formula for comprehensively evaluating the storage life of complex electromechanical products is as follows:

[0035]

[0036] Given any given moment, the probability competition weight formula is used to obtain the probability competition weights of each newly produced product, product weak link and iterative product, and then the weight of each product and the storage life of each product corresponding to the reliability are substituted into the probability weighted model formula of the complex electromechanical product life to obtain the estimated storage life of the electromechanical product. The estimated storage life is taken as equal to the storage life of the electromechanical product with the calculation result corresponding to any given moment and the reliability R. At this time, the given storage period is consistent with the estimated value of the reliable storage life of the complex electromechanical product. The storage life of the complex electromechanical product is equivalent to the probability competition weight of the storage life of each type of product under the given reliability R.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] (1) The present invention comprehensively considers the reliability of newly produced products, product weaknesses and iterative products, and solves the problem that the traditional storage life assessment method of complex electromechanical products only considers a single type of product, ignores the impact of newly produced products, iterative products and product weaknesses on the storage life assessment of complex electromechanical products, resulting in the difficulty in ensuring the accuracy of the storage life assessment of complex electromechanical products. The present invention is particularly suitable for the storage life assessment of products that have not yet reached their lifespan and are newly finalized with high reliability and long life.

[0039] (2) The present invention improves the accuracy of storage life assessment of complex electromechanical products. The probability competition weight is introduced to establish a probability weighted model for the life of complex electromechanical products. The probability competition weight formula is constructed by cleverly combining the reliability functions of new products, product weaknesses, and iterative products. This can effectively improve the accuracy of storage life assessment of complex electromechanical products with high reliability and long life. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a flow chart of the comprehensive evaluation method of storage life of complex electromechanical products based on life probability weighting of the present invention;

[0041] Figure 2 It is a schematic diagram of a stress profile of a comprehensive evaluation method for storage life of complex electromechanical products based on life probability weighting of the present invention;

[0042] Figure 3 It is a reliability function curve diagram of a newly produced electromechanical fuse of a certain type according to the comprehensive evaluation method of storage life of complex electromechanical products based on life probability weighting of the present invention;

[0043] Figure 4 A reliability function curve diagram of a certain type of electromechanical fuze power module according to the comprehensive evaluation method for storage life of complex electromechanical products based on life probability weighting of the present invention;

[0044] Figure 5 It is a reliability function curve diagram of an iterative product of a certain type of electromechanical fuze according to the comprehensive evaluation method of storage life of complex electromechanical products based on life probability weighting of the present invention;

[0045] Figure 6 This is a diagram of the storage life solution process of a certain type of electromechanical fuze based on the comprehensive storage life evaluation method of complex electromechanical products based on life probability weighting of the present invention. DETAILED DESCRIPTION

[0046] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0047] like Figure 1 As shown, the embodiment of the present application provides a comprehensive evaluation method for storage life of complex electromechanical products based on life probability weighting, comprising the following steps:

[0048] A1: Obtain test data of newly produced products, weak links of products and iterative products, run the stepped temperature stress accelerated life test according to the set stress level value, and obtain the life data of each product to be evaluated;

[0049] Among them, the weak link of the product is the weak component of the product. For example, the electromechanical fuze includes the detection and detonation control module, the safety and detonation module, the power module, etc. If the power module is the weakest component of the fuze, the life of the weak component is taken as the life of the product (i.e. the barrel theory). The iterative product refers to the previous generation of the product with the required life. For fuzes and ammunition, the replacement is carried out on the basis of the previous generation of products. For example, the second generation product improves the output power of the power supply on the basis of the first generation product.

[0050] Accelerated life test uses the life characteristics of products under high stress to extrapolate the life characteristics under normal stress, and the Arrhenius model is the most typical and widely used accelerated model when using temperature stress test. It is an empirical formula for the relationship between the chemical reaction rate constant and temperature created by Swedish Arrhenius. The life data of the products to be evaluated include the life data of new products, the life data of weak links of products, and the life data of iterative products.

[0051] Furthermore, obtaining test data for newly produced products and iterative products includes: conducting performance tests on the performance of newly produced products and iterative products, judging product failures based on product manufacturing and factory acceptance specifications, and obtaining life data for each product; the performance indicators include release resistance, ignition resistance, high, low and high explosion release times, release amplitudes, ignition amplitudes, etc.

[0052] Furthermore, obtaining test data of weak links of the product includes: performing FMECA analysis on each module of the product and the components of the module, and then performing stress strengthening test on each module of the product to obtain module data corresponding to the weak links of the product.

[0053] This FMECA targets all possible failures of the product and, based on the analysis of the failure modes, determines the impact of each failure mode on the product operation, identifies single point failures, and determines the harmfulness of the failure mode according to its severity and probability of occurrence.

[0054] Furthermore, if Figure 2 As shown, running the stepped temperature stress accelerated life test according to the set stress level value includes: first selecting a set of accelerated stress levels, and at the beginning of the test, placing a set number of samples at the first stress level S 1 Accelerated tests are carried out under the condition of step temperature stress acceleration test; and product performance parameters are measured by m 1 Test once; after acceleration time t 1 Then increase the stress level to the second stress level S 2 , place the unfailed samples at the second stress level and continue the life test. 2 -t 1 ) during which samples were taken out for m 2Performance test: increase the stress level in sequence according to the step stress accelerated life test, and stop the test when the test cut-off time is reached. 1 ,S 2 ,…,S k}, the normal stress level is S 0 <S 1 <S 2 <…<S k , S 0 , stress is the step temperature. The sample is the product, the object of the test, and also a complex electromechanical product.

[0055] The stress acceleration test plan is formulated in accordance with the relevant provisions of the relevant gjb (national military standard) to formulate an accelerated test plan suitable for the product. For example, gjb5103 stipulates that the number of test samples for the step stress acceleration test shall not be less than 12, and the stress level shall not be less than 4, etc. A test plan suitable for the product is formulated according to the gjb.

[0056] A2: According to the type of life data of the product to be evaluated, the product characteristic life is obtained by combining the Arrhenius acceleration model and the Weibull distribution. The model parameters are estimated using the maximum likelihood estimation method and the least squares method. The maximum likelihood estimation of the shape parameters and the product characteristic life under normal stress is obtained, and the storage life of the product under normal stress level is evaluated; the reliability function and storage life of the newly produced products, product weak links, and iterative products (life data of the product to be evaluated) are obtained.

[0057] Furthermore, the Arrhenius acceleration model formula is as follows:

[0058]

[0059] Among them, η i , i=1,2,...,l is the characteristic life of the product, l is the number of stress levels; a,b are estimated parameters; Ti is the absolute temperature. Since formula (1) is a simplified formula of the Arrhenius acceleration model, the original formula is A is an unknown constant related to the experiment; E is the activation energy; K is the Boltzmann constant, which is a fixed value, 8.6×10 -5 eV / K. Later, the Arrhenius acceleration model was used in the step stress accelerated life test and became a simplified form. After simplification, a and b are estimated parameters.

[0060] The Weibull distribution fitting calculation is performed on the product characteristic life of the random sample, and the product characteristic life is used as the scale parameter of the Weibull distribution. Assuming that the product characteristic life distribution obeys the Weibull distribution, the failure distribution function is as follows:

[0061]

[0062] Among them, m i is the shape parameter, η i (eta) is the scale parameter (product characteristic life) under normal stress level. Weibull distribution has shape parameter and scale parameter. The shape parameter is used to describe the shape characteristics of probability distribution, which can affect the kurtosis and skewness of the distribution curve. In practical applications, Weibull distribution is often used to describe the duration and survival time of random events. For example, for product life analysis, Weibull distribution is used to estimate the failure time of the product. In addition, Weibull is used in the field of reliability and risk analysis.

[0063] Furthermore, according to the stress level set in the accelerated test and combined with the Weibull distribution, the failure distribution function formula after time conversion at each set stress level is obtained as follows:

[0064]

[0065] The failure distribution function after time conversion under the set stress level includes: based on the basic assumptions, the conversion distribution of the product storage life under the step stress accelerated test conditions is derived. The specific derivation process is:

[0066] First, the sample is heated to absolute temperature T 1 The sample test is carried out under the condition of t 1 ; then raise the temperature to T 2 , the test time is t 2 -t 1 ; then in T 3 Next test t 3 -t2, in T 4 Next test t 4 -t 3 , and finally in T l Next test t l -t l-1 , the test is at t l The moment is over.

[0067] According to the basic assumption, in the time period [0,t 1 The lifetime distribution of the samples on ] is:

[0068]

[0069] At time (t 1 ,t 2 ], when in T 2 When the test was carried out under T 1 A period of time has been carried out, which cannot be ignored, so this period of time is converted into T 2 Set the time to S1 , we get 1 ,t 2 The lifetime distribution of the samples on ] is:

[0070] F(t)=F 2 (S 1 +(tt 1 )), t 1 <t≤t 2

[0071] According to the basic assumption: F 1 (t 1 )=F 2 (S 1 )

[0072] Therefore, substituting the above formula into the life distribution, we get:

[0073]

[0074] Similarly, there is S 2 Satisfaction: F 3 (S 2 )=F 2 (S 1 +(t 2 -t 1 )),

[0075] And F(t)=F 3 (S 2 +(tt 2 )),t 2 <t≤t 3 ,

[0076] Right now

[0077] After similar operations, we get:

[0078]

[0079] The above calculation results are expressed using a failure distribution function formula:

[0080]

[0081] After conversion, the failure distribution function formula corresponding to the temperature step stress accelerated test is obtained. The product life data of the accelerated test is of success or failure type and can be represented by binomial distribution. Therefore, the likelihood function of the sample is:

[0082]

[0083] The cumulative distribution function refers to the probability that a product will fail under specified conditions and within a specified time, denoted as F(t). The cumulative distribution function can be expressed by the following formula:

[0084]

[0085] As we all know, F(t) and R(t) are a pair of complementary functions, that is, R(t)+F(t) = 1. Because at the beginning of work, no product has any problems, that is, R(0) = 1, F(0) = 0. As time goes by, the number of failures also increases, and reliability decreases accordingly.

[0086] Furthermore, the maximum likelihood estimation method is used to find the shape parameter m and the scale parameter η of the log-likelihood function. i The partial derivative of the shape parameter and scale parameter are obtained; then according to the Arrhenius acceleration model formula and the scale parameter η i The estimated values ​​of parameters a and b are obtained by the least squares method; where the partial derivative is set to 0.

[0087] Furthermore, the maximum likelihood estimation of the shape parameters and the characteristic life of the product under normal stress includes: obtaining the likelihood function of the life data under each stress according to the failure distribution function converted by the reference time, the number of samples and the number of failures:

[0088]

[0089] The log-likelihood function of the life data under each stress is obtained as follows:

[0090]

[0091] Among them, n i is the stress level S i The number of samples under r i is the stress level S i The number of failures below.

[0092] Furthermore, the least squares estimation model is used to calculate the estimated values ​​of parameters a and b, and the logarithms of both sides of the Arrhenius acceleration model formula (1) are taken:

[0093]

[0094] Let y = lnη i , Get the estimated values ​​of parameters a and b.

[0095] Furthermore, the estimated values ​​of shape parameters and scale parameters are obtained according to the estimation model to evaluate the characteristic life under normal stress level; 0 The characteristic lifetime η under 0 Estimated value of for:

[0096]

[0097] Specifically, for product storage life assessment, the reliability function under normal storage conditions is:

[0098]

[0099] The reliability function formula (4) gives the storage life of a product with a given reliability under normal storage conditions. R(t) is the function of the reliability R with respect to time t, m 0 is the shape parameter at normal stress level, η 0 (eta) is the scale parameter (product characteristic life) at normal stress level.

[0100] Furthermore, the combination of the Arrhenius acceleration model and the Weibull distribution includes: under different stress levels, the product failure mechanism is the same, that is, under each stress, the shape parameter in the life distribution remains unchanged, that is, m 0 =m 1 =…=m l .

[0101] A3: Establish a probability weighted model for the life of complex electromechanical products, calculate the product probability competition weights, and comprehensively evaluate the storage life of complex electromechanical products.

[0102] Furthermore, the establishment of a weighted probability model for the life of complex electromechanical products includes: obtaining the reliability function of newly produced products, weak links and iterative products, obtaining the storage life of each type of newly produced products, product weak links and iterative products under a given reliability according to the reliability function, and using the storage life and reliability function to calculate the probability competition weight of each product to obtain a comprehensive evaluation of the storage life of complex electromechanical products. Specifically, the given reliability range is different according to different product quality requirements, and the given reliability is not less than 0.9. By comprehensively utilizing the historical information of the equipment during long-term use, a storage reliability function model of the whole machine is established. Because the storage life value is required to evaluate the comprehensive life, in order to substitute it into the reliability function, a reliability R=0.9 is given to calculate the storage life t of the product.

[0103] Furthermore, the formula for establishing a life probability weighted model for complex electromechanical products is:

[0104]

[0105] Among them,R,j For j At the moment, the storage life of complex electromechanical products when the reliability is R; n is the number of products of each type to be weighted, ω i,j is the probability competition weight, that is, at t j The normalized value of the failure probability of the i-th type of product at time T R,i is the storage life of the i-th type of product when the reliability is R. In the present invention, the reliability and storage life of newly produced products, product weaknesses and iterative products are comprehensively considered, so n is 3.

[0106] Further, calculate the product probability competition weight ω i,j The formula is as follows:

[0107]

[0108] Among them, R i (t j ) is at t j The storage reliability of the i-th type of product at the moment is used to calculate the product probability competition weight based on the reliability function R(t).

[0109] Furthermore, the formula for comprehensively evaluating the storage life of complex electromechanical products is as follows:

[0110]

[0111] The above formula for the storage life of complex electromechanical products is the probability competition weight formula, which calculates the comprehensive storage life through the probability competition weight and storage life; the formula below is the actual storage life formula of the product. The two formulas together indicate that the actual storage life of the product is the same as the life after adopting the life probability weighting. Given any time t j , use the probability competition weight formula (6) to obtain the probability competition weight of each new product, product weak link and iterative product, and then substitute the probability competition weight of each product and the storage life of each product corresponding to the reliability into the probability weighted model formula (5) of the complex electromechanical product life to obtain the estimated value of the comprehensive storage life of the electromechanical product Take the estimated storage life equal to any given The calculation result corresponding to the moment is the storage life of the electromechanical product reliability R. At this time, the given storage years are consistent with the estimated value of the reliable storage life of the complex electromechanical product. The storage life of the complex electromechanical product is equivalent to the probability competition weight of the storage life of each type of product under the given reliability R, where the probability competition weight is the normalized value of the failure probability of each type of product under the storage years.

[0112] Therefore, under a given reliability R, The fitting curve and The abscissa of the intersection of the two points is used to comprehensively evaluate the storage life of complex electromechanical products (such as Figure 6 ).

[0113] Taking a certain type of electromechanical fuze as an example, the present invention is described as follows:

[0114] Step 1: Run the stepped temperature stress accelerated life test for new products, weak links of products, and iterative products to obtain the life data of each product;

[0115] This example uses three types of products, namely a newly produced electromechanical fuze, a power module with weak storage links in an electromechanical fuze, and an iterative product of an electromechanical fuze, as research objects to carry out storage life tests. The accelerated test plan and test data are as follows:

[0116] (1) A new type of electromechanical fuze

[0117] 1) Test conditions

[0118] A step-by-step temperature stress accelerated life test was conducted for a newly produced electromechanical fuze. Four stress levels were set, the test sample size was 48 rounds, and the total test time was 153 days. The test was conducted at a temperature stress level of 80°C for 66 days (4 tests in total), 85°C for 48 days (6 tests in total), 90°C for 33 days (8 tests in total), and 95°C for 6 days (3 tests in total). The temperature control deviation was ±2°C.

[0119] 2) Test data

[0120] The performance of the electromechanical fuse release resistance, ignition resistance, high and low explosion height release time, release amplitude, ignition amplitude, etc. is tested. The product failure is judged according to the product manufacturing and factory acceptance specifications to form the fuze life data. The data of a newly produced electromechanical fuze of the step stress accelerated life test is shown in Table 1.

[0121] Table 1

[0122]

[0123]

[0124] (2) The weak link of a certain type of electromechanical fuze: the power module

[0125] A certain type of electromechanical fuze is divided into a detection and detonation control module, a safety and detonation module, and a power module. The failure modes, impacts, and criticality of each module and component during storage are analyzed (FMECA analysis). The FMECA analysis results show that the power module of the electromechanical fuze is its weak link in storage. The magnetic core of the power module generates magnetic field concentration, resulting in no performance output, which eventually causes the fuze to misfire. The coil has no performance output due to short circuit or open circuit, which eventually causes the fuze to misfire. In addition, stress strengthening tests were carried out on each module of the fuze. In the test, the power module of the fuze failed first. The fuze was analyzed by two analysis methods, FMECA analysis and stress strengthening test. The results show that the weak link of the fuze is the power module.

[0126] 1) Test conditions

[0127] The power module consists of a turbine generator, a power circuit, etc. Its function is to provide working power for the detection and detonation control module. A step stress accelerated life test was conducted on the module, with four stress levels set in total. The test sample size was 48 rounds, and the total test time was 153 days. The test was conducted at a temperature stress level of 80°C for 66 days (4 tests in total), 85°C for 48 days (6 tests in total), 90°C for 33 days (8 tests in total), and 95°C for 6 days (3 tests in total). The temperature control deviation was ±2°C.

[0128] 2) Test data

[0129] The power module's 12V output voltage average value, 0V to 12V rise time, square wave peak-to-peak value, etc. are tested for performance. The product failure is determined according to the product manufacturing and factory acceptance specifications to form the power module life data. The power module data of the step temperature stress accelerated life test is shown in Table 2.

[0130] Table 2

[0131]

[0132]

[0133] (3) Iteration product of a certain type of electromechanical fuze

[0134] A step stress accelerated life test was conducted on a certain type of electromechanical fuze iteration product. Four stress levels were set, the test sample size was 50 rounds, and the total test time was 150 days. The test was conducted at a temperature stress level of 75°C for 60 days (3 tests in total), 80°C for 45 days (3 tests in total), 85°C for 30 days (3 tests in total), and 90°C for 15 days (3 tests in total). The temperature control deviation was ±2°C.

[0135] 2) Test data

[0136] The performance tests are carried out on the release resistance, ignition resistance, release time of high and low explosion height, release amplitude, ignition amplitude, etc. of the electromechanical fuze; the product failure is judged according to the product manufacturing and factory acceptance specifications to form the fuze life data. The iterative product data of a certain type of electromechanical fuze in the step temperature stress accelerated life test is shown in Table 3.

[0137] Table 3

[0138]

[0139]

[0140] Step 2: Combining the Arrhenius acceleration model and Weibull distribution, the maximum likelihood estimation and least squares method are used to estimate the model parameters and evaluate the shelf life of the product.

[0141] (1) Storage life assessment of a newly produced electromechanical fuze

[0142] Based on the step temperature stress accelerated life test data of a newly produced electromechanical fuze, the maximum likelihood estimation and least squares method are used to estimate the model parameters by combining the Arrhenius acceleration model and the Weibull distribution. The maximum likelihood estimates of the shape parameters and characteristic life are obtained as follows: 0 =15.9636,η 0 =7301.5004. The reliability function under normal stress level is:

[0143]

[0144] from Figure 3 It can be seen that when the reliability of the fuze is 0.9, the reliable life of the newly produced product is 6334 days, or 17.3534 years.

[0145] (2) Storage life assessment of the power module, a weak link in a certain type of electromechanical fuze

[0146] Based on the power module step temperature stress accelerated life test data, combined with the Arrhenius acceleration model and Weibull distribution, the model parameters are estimated using maximum likelihood estimation and least squares method, and the maximum likelihood estimates of shape parameters and characteristic life are obtained as follows: m 0 =16.7496,η 0 =6975.1028. The reliability function under normal stress level is:

[0147]

[0148] from Figure 4 It can be seen that when the reliability of the power module is 0.9, the reliable life of the product is 6091 days, or 16.6876 years.

[0149] (3) Storage life assessment of a certain type of electromechanical fuze iteration product

[0150] Based on the stepped temperature stress accelerated life test data of a certain type of electromechanical fuze iteration product, the maximum likelihood estimation and least squares method are used to estimate the model parameters by combining the Arrhenius acceleration model and the Weibull distribution. The maximum likelihood estimates of the shape parameters and characteristic life are: m 0 =16.8737,η 0 =6039.2973. The reliability function under normal stress level is:

[0151]

[0152] from Figure 5 It can be seen from the reliability function curve that when the reliability of the fuze is 0.9, the reliable life of the product is 5290 days, or 14.4932 years.

[0153] Step 3: Establish a probability weighted model for the life of complex electromechanical products to comprehensively evaluate the product storage life.

[0154] The reliable life of each type of product under normal stress is obtained by formula (12), formula (13) and formula (14). When the reliability is given as 0.9, the evaluation results of the reliable storage life of each type of product are shown in the following table.

[0155] Table 4

[0156]

[0157] Note: Lifespan is measured in years.

[0158] Given a reliability of R = 0.9, the reliable storage life of the system is:

[0159]

[0160] in, ω i (t) is the probability competition weight, that is, the normalized value of the failure probability of the i-th type of product when the storage time is t, and the expression is:

[0161]

[0162] Substituting formula (12), formula (13), and formula (14) into formula (16), we can get ω A (t),ω B (t),ω C (t).

[0163] Combine formula (15) with Combined to get:

[0164]

[0165] Draw the function graph of formula (17), and the horizontal coordinate of the intersection is the reliable life of the electromechanical fuze. Figure 6 It can be seen that the reliable life of the electromechanical fuze is 5411 days, or 14.8247 years.

[0166] Substituting the reliable storage life of a certain type of electromechanical fuze, 5411 days, into formula (16), the probability competition weights of various types of products can be obtained, and the results are shown in Table 5. As can be seen from the table, when the reliability is 0.9, the life of the iterative product has the highest weight on the evaluation result of the storage life of the electromechanical fuze.

[0167] Table 5

[0168] Reliability <![CDATA[ω A (t)]]> <![CDATA[ω B (t0]]> <![CDATA[ω C (t)]]> Reliable storage life of a certain type of electromechanical fuze / day (year) 0.9 0.0498 0.0843 0.8659 5411(14.8247)

[0169] In the description of the present invention, it is necessary to understand that the indicated orientation or positional relationship is based on the orientation or positional relationship shown in the accompanying drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0170] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A comprehensive evaluation method for storage life of complex electromechanical products based on life probability weighting, characterized by: include: Obtain the indicator data of newly produced products, product weaknesses and iterative products, run the stepped temperature stress accelerated life test according to the selected stress level value, and obtain the life data of each product to be evaluated; According to the life data type of the product to be evaluated, the product characteristic life is obtained by combining the Arrhenius acceleration model and the Weibull distribution. The model parameters are estimated using the maximum likelihood estimation method and the least squares method. The maximum likelihood estimation of the shape parameters and the product characteristic life under normal stress is obtained, and the storage life of the product under normal stress level is evaluated respectively. Obtain the reliability function and storage life of new products, product weaknesses, and iterative products; Establish a probability weighted model for the life of complex electromechanical products, calculate the product probability competition weights, and comprehensively evaluate the storage life of complex electromechanical products.

2. The method according to claim 1, characterized in that: The step temperature stress accelerated life test according to the selected stress level value includes: first selecting a group of accelerated stress levels, placing a set number of samples at a first stress level for accelerated testing at the beginning of the test; and testing product performance parameters; increasing the stress level to a second stress level after the acceleration time, placing the samples that have not failed at the second stress level to continue the life test, and taking out the samples for performance testing during the test; increasing the stress level in sequence according to the step stress accelerated life test, and stopping the test when the cut-off time specified in the test is reached.

3. The method according to claim 1, characterized in that: The Arrhenius acceleration model formula is as follows: Among them, η i ,i=1,2,...,l is the characteristic life of the product, l is the number of stress levels; a, b are estimated parameters; Ti is the absolute temperature.

4. The method according to claim 1, characterized in that: Combined with the Weibull distribution, the failure distribution function formula after time conversion under each set stress level is as follows: Among them, m i is the shape parameter, η i is the scale parameter at normal stress level.

5. The method according to claim 1, characterized in that: The maximum likelihood estimation method is used to calculate the partial derivatives of the shape parameter and the scale parameter of the log-likelihood function to obtain the estimated values ​​of the shape parameter and the scale parameter; and then the estimated value of the parameter is obtained by the least squares method according to the Arrhenius acceleration model formula and the estimated value of the scale parameter.

6. The method according to claim 5, characterized in that: The maximum likelihood estimation of the shape parameters and the characteristic life of the product under normal stress includes: obtaining the likelihood function of the life data under each stress according to the failure distribution function converted from the reference time, the number of samples and the number of failures: Find the logarithm of the likelihood function.

7. The method according to claim 6, characterized in that: The characteristic life under normal stress level is evaluated based on the estimated values ​​of the shape parameters and scale parameters of the estimated model.

8. The method according to claim 1, characterized in that: The establishment of a probability weighted model for the life of complex electromechanical products includes: obtaining reliability functions of newly produced products, weak links and iterative products, and then obtaining the storage life of various types of newly produced products, product weak links and iterative products based on a given reliability function, and using the storage life and reliability function to calculate the probability competition weight of each product to obtain a comprehensive evaluation of the storage life of the complex electromechanical products.

9. The method according to claim 1, characterized in that: The formula for establishing a probability weighted model for the life of complex electromechanical products is: Among them, R,j For j At the moment, the storage life of complex electromechanical products when the reliability is R; n is the number of products of each type to be weighted, ω i,j is the probability competition weight, that is, at t j The normalized value of the failure probability of the i-th type of product at time T R,i is the storage life of the i-th type product when the reliability is R.

10. The method according to claim 1, characterized in that: The formula for comprehensively evaluating the storage life of complex electromechanical products is as follows: Given any given moment, the probability competition weight formula is used to obtain the probability competition weights of each newly produced product, product weak link and iterative product, and then the weight of each product and the storage life of each product corresponding to the reliability are substituted into the probability weighted model formula of the complex electromechanical product life to obtain the estimated storage life of the electromechanical product. The estimated storage life is taken as the storage life of the electromechanical product with the calculation result corresponding to any given moment as the reliability. At this time, the given storage period is consistent with the estimated value of the reliable storage life of the complex electromechanical product. The storage life of the complex electromechanical product is equivalent to the probability competition weight of the storage life of each type of product under a given reliability R.

Images