A multi-element step stress accelerated degradation modeling and reliability evaluation method, device and medium

Abstract

The application provides a multi-element step stress accelerated degradation modeling and reliability evaluation method, equipment and medium, comprising: obtaining step stress accelerated degradation data of a target product to generate a data set; establishing a multi-element step stress accelerated degradation model; fitting the data set by using the multi-element step stress accelerated degradation model to estimate model parameters; and calculating the product reliability of the target product under normal use conditions based on the estimated model parameters and the multi-element step stress accelerated degradation model. The application has the characteristics of clear model form, simple parameter estimation and easy application.

Description

Technical Field

[0001] This invention relates to the fields of reliability engineering and lifetime prediction technology, and in particular to a method, device and medium for multivariate step stress accelerated degradation modeling and reliability assessment. Background Technology

[0002] With the development of technology, many products, especially military electronic products, often have high reliability and long service life, making it difficult to conduct life tests and obtain failure data. Therefore, traditional reliability assessment methods based on life tests are no longer applicable. Degradation testing, and even accelerated degradation testing, is gradually becoming the mainstream method for product reliability assessment due to its shorter testing time and lower cost.

[0003] The fundamental theory of accelerated degradation testing is the cumulative damage model proposed by Nelson et al. [Nelson W. Accelerated lifetesting step-stress models and data analyses [J]. IEEE Transactions on Reliability 1980; 29(2): 103-108.], which assumes that the degradation process of a product is only related to the current environmental stress. Generally, accelerated degradation testing can be subdivided into constant stress accelerated degradation testing and step stress accelerated degradation testing [Huang J, Golubovi...]. DS, Koh S, Yang D, Li X, Fan X, Zhang G Q. Lumen degradation modeling of white-light LEDs in step stress accelerated degradation test [J]. Reliability Engineering & System Safety 2016; 154: 152-159. In constant stress accelerated degradation tests, a set of test samples needs to be placed at each stress level. However, in step stress accelerated degradation tests, the same set of test samples can be tested sequentially at different stress levels. Therefore, compared to constant stress accelerated degradation tests, step stress accelerated degradation tests have a lower cost-effectiveness ratio. Currently, step stress accelerated degradation tests are widely used, and step stress accelerated degradation models have also received widespread attention and research.

[0004] Before introducing this invention, we will first review the current research status of step stress accelerated degradation models both domestically and internationally: Currently, research on accelerated degradation models using step stress is relatively comprehensive. For example, Wang et al. [Wang L, Pan R, LiX, and Jiang T. A Bayesian reliability evaluation method with integrated accelerated degradation testing and field information [J]. Reliability Engineering & System Safety 2013; 112: 38-47.] proposed a Bayesian method to integrate laboratory accelerated degradation data with field failure data, thereby achieving product reliability assessment. Tang et al. [Tang S, Guo X, Yu C, Xue H, and Zhou Z. Accelerated degradation tests modeling based on the nonlinear Wiener process with random effects [J]. Mathematical Problems in Engineering 2014; 2014: 1-11.] proposed a nonlinear Wiener process model considering random effects of parameters to model accelerated degradation data using step stress. In this model, the drift coefficient of the Wiener process is represented as the product of a random variable and a function of stress. Huang et al. [Huang J et al. Lumendegradation modeling of white-light LEDs in step stress accelerated degradation test [J]. Reliability Engineering & System Safety 2016; 154: 152-159.] proposed a modified Brownian motion to describe the light intensity degradation process of LEDs under step stress accelerated degradation test conditions.Li et al. [Li J, Wang Z, Liu X, Zhang Y, Fu H, and Liu C. A Wienerprocess model for accelerated degradation analysis considering measurement errors [J]. Microelectronics Reliability 2016; 65: 8-15.] proposed a Wiener process model considering measurement errors for step stress accelerated degradation data. Building upon this, Li et al. [Li J, Wang Z, Liu C, and Qiu M. Stochastic accelerated degradation model involving multiple accelerating variables by considering measurement error [J]. Journal of Mechanical Science and Technology 2019; 33(11): 5425-5435.] further proposed a step stress accelerated degradation model considering multiple accelerating stresses. Li et al. [Li Y, Gao H, Chen H, Liu C, Yang Z, and Zio E. Accelerated degradation testing for lifetime analysis considering random effects and the influence of stress and measurement errors [J]. Reliability Engineering & System Safety 2024; 247: 110101.] and Yang et al. [Yang C, Gu X, and Xiao Z. Reliability estimation from two types of accelerated testing considering individual difference and measurement error [J]. Probabilistic Engineering Mechanics 2024; 75: 103584.] proposed a step stress accelerated degradation model that simultaneously considers the random effects of parameters and measurement errors based on the Wiener process.Pang et al. [Pang Z, Hu C, Si X, Zhang J, Du D, and Pei H. Nonlinear step-stress accelerated degradation modeling and remaining useful life estimation considering multiple sources of variability [J]. IEEE Access 2019; 7: 124558-124575.] and Feng et al. [Feng X, Tang J, and Balakrishnan N. Reliability inference for dual stress factors accelerated degradation test based on the nonlinear Wiener process with three-source variability [J]. Communications in Statistics - Theory and Methods 2025; 54(21): 6866-6889.] extended the Wiener over-model considering three-source uncertainties to a step-stress accelerated degradation model. For monotonic degradation data, Jiang et al. [Jiang P, Wang B, Wang X, and Tsai TR. Reliability assessment and remaining useful life prediction based on the inverse Gaussian step-stress accelerated degradation data [J]. IEEE Transactions on Reliability 2024; 73(2): 967-977.] proposed a step-stress accelerated degradation model based on an inverse Gaussian process, in which the shape and scale parameters of the inverse Gaussian process are assumed to be functions of stress.To address the problem of outliers, Li et al. [Li Y, Fei M, Jia L, Lu N, Kaynak O, and Zio E. Novel outlier-robust accelerated degradation testing model and lifetime analysis method considering time-stress-dependent factors. IEEE Transactions on Industrial Informatics 2024; 20(8): 9907-9917.] proposed a step stress-accelerated degradation model based on the Wiener process, in which the random effects in the drift coefficients are assumed to follow a t-distribution rather than a traditional normal distribution.

[0005] Although numerous accelerated degradation models based on step stress have been proposed in the literature, most only consider the degradation of a single performance metric. However, for many complex products, testing requires monitoring the degradation data of multiple performance metrics simultaneously, rather than focusing on a single metric. For example, for the T / R assembly, a core component of a phased antenna array, it is often necessary to simultaneously monitor multiple performance metrics such as output channel power, receive channel gain, and amplitude-phase consistency across multiple channels. In accelerated degradation tests based on step stress, these different performance metrics may exhibit degradation trends simultaneously. Therefore, it is necessary to establish a multivariate accelerated degradation model based on step stress to achieve effective fitting of multidimensional degradation data and accurate assessment of product reliability. To address the aforementioned needs, only a few studies have considered the joint degradation of multiple performance indices when establishing step stress accelerated degradation models. For example, Sun et al. [Sun F, Liu L, Li X, and Liao H. Stochastic modeling and analysis of multiple nonlinear accelerated degradation processes through information fusion [J] Sensors 2016; 16(8): 1242.] proposed a multivariate step stress accelerated degradation model based on the Wiener process and copula function. However, the parameter estimation of this model is quite complex, and the selection of the optimal copula function is also difficult. Lu et al. [Lu L, Wang B, Hong Y, and Ye Z. General path models for degradation data with multiple characteristics and covariates [J]. Technometrics 2021; 63(3): 354-369.] proposed a multivariate degradation trajectory model considering covariates, which showed satisfactory fitting results on the three-dimensional degradation dataset of coatings. However, their model cannot be directly applied to multivariate step stress accelerated degradation data. Summary of the Invention

[0006] This application provides a method, device, and medium for multi-dimensional step stress accelerated degradation modeling and reliability assessment to solve the data modeling problem of multi-performance index products in step stress accelerated degradation tests.

[0007] Other features and advantages of this application will become apparent from the following detailed description, or may be learned in part from practice of this application.

[0008] According to a first aspect of the embodiments of this application, a method for multivariate step stress accelerated degradation modeling and reliability assessment is provided, including: Acquire step stress-accelerated degradation data and generate a dataset; Establish a multivariate step stress accelerated degradation model; The dataset was fitted using the aforementioned multivariate step stress accelerated degradation model to estimate the model parameters; The reliability of the target product under normal use conditions is calculated based on the estimated model parameters and the multivariate step stress accelerated degradation model.

[0009] According to one embodiment of this application, the step of acquiring accelerated degradation data by step stress and generating a dataset specifically includes: The number of samples participating in the experiment was set as follows: n The number of performance indicators for each sample is k The number of accelerated stress levels is l In the q The number of degradation observations at each accelerated stress level is , q = 1, 2, …, l ; Each sample was placed sequentially l In each accelerated stress level, the accelerated stress level increases sequentially, and the observation time is divided for each accelerated stress level to obtain the measured value of degradation of each sample at each observation time. The measured value of degradation consists of the performance index of at least one product. The dataset consists of the observation time, accelerated stress level, and measured degradation value for each sample.

[0010] According to one embodiment of this application, establishing the multivariate step stress accelerated degradation model specifically includes: Assuming the observation time In the first u Acceleration stress level Below, that is , u Values ​​range from 1 to l The degradation observations are modeled as follows:

[0011] in, , is the stress influence coefficient matrix, which contains all k The performance index at the first u ( u = 1, 2, …, l The stress influence coefficients at each stress level are: The first in p (p = 1, 2, …, k ) diagonal elements Indicates the first p The performance index at the first u Acceleration stress level Stress influence coefficient under the following conditions Similarly; It is the baseline cumulative damage matrix, which contains all k Each performance metric has a cutoff time. Baseline cumulative damage at time, where, The first in p ( p = 1, 2,…, k ) diagonal elements Indicates the first p Each performance metric has a cutoff time. Baseline cumulative damage at time Similarly; It is a random effects vector, which contains all k The heterogeneity of the degradation trajectory of each performance index among different samples, where, for The first in p ( p = 1,2, …, k ) elements, For fixed parameters, To obey k Random parameters of a dimensional normal distribution; It is a random error vector, representing the error at the observation time. The first to last fit k Unexplained fitting errors that arise during the degradation process of a performance metric.

[0012] According to one embodiment of this application, the accelerated degradation model is any one of the Arrhenius model, the inverse power law model, and the exponential model.

[0013] According to one embodiment of this application, the estimated model parameters specifically include: For the obtained dataset is , Let be the measured value of the degradation of the i-th sample at the j-th observation time. The constructed multivariate step stress-accelerated degradation model is expressed as:

[0014] in, , , , ; in, The parameters in the model are often estimated using the multivariate least squares method. The parameters are obtained using maximum likelihood estimation.

[0015] According to one embodiment of this application, the The parameters in the model are commonly estimated using the multivariate least squares method, specifically including: Define the sum of squared errors (SSE) as follows:

[0016] SSE pair Find the first derivative and set it to zero, then we get... estimator :

[0017] estimate Substituting the algebraic expression into the SSE formula and minimizing the SSE, we obtain... The estimation results; Then redo Substituting into the SEE formula, we get Numerical results; based on The numerical results were obtained by using maximum likelihood estimation to obtain the parameters. and The estimate.

[0018] According to one embodiment of this application, the The parameters in the model are obtained using maximum likelihood estimation, and specifically include: Fitting error variance The maximum likelihood estimate of the covariance matrix is

[0019] matrix The square root of the diagonal element is used as The estimate.

[0020] According to one embodiment of this application, the formula for calculating the product reliability is as follows:

[0021] in, This refers to the stress level of the target product under normal operating conditions.

[0022] According to a second aspect of the embodiments of this application, an electronic device is provided, comprising: At least one processor; and a memory communicatively connected to said at least one processor; The memory stores instructions that can be executed by the at least one processor, and the at least one processor executes the instructions stored in the memory to perform the method described in the first aspect.

[0023] According to a third aspect of the embodiments of this application, a computer-readable storage medium is provided for storing instructions that, when executed, cause the method described in the first aspect to be implemented.

[0024] Compared with existing technologies, the advantages of adopting the above technical solution are as follows: This invention performs statistical modeling on data collected from typical accelerated degradation tests of multi-performance index products. The model considers both the heterogeneity between samples and the correlation between performance indexes, and also introduces the influence of accelerated stress level on the degradation process. It features a clear model form, simple parameter estimation, and ease of application. Attached Figure Description

[0025] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. It is obvious that the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0026] Figure 1 This is a flowchart illustrating the multi-step stress accelerated degradation modeling and reliability assessment method according to an embodiment of this application.

[0027] Figure 2 This is a graph showing the relationship between stress level and observation time in the step stress accelerated degradation test targeted by the embodiments of this application.

[0028] Figure 3 This is a simulation of the degradation data of the first performance index of the product obtained in the simulation test of the embodiment of this application, as well as a schematic diagram of the three temperature stress levels at which it is located.

[0029] Figure 4 This is a schematic diagram showing the fitting results of the multivariate step stress accelerated degradation model proposed in this application to the simulated degradation data. (5 out of 10 samples were randomly selected as examples for demonstration).

[0030] Figure 5 This is a schematic diagram illustrating the reliability assessment results of the product under normal operating stress conditions according to an embodiment of this application.

[0031] Figure 6This is a schematic diagram of the structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0032] The embodiments of this application are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar modules or modules having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application. Rather, the embodiments of this application include all variations, modifications, and equivalents falling within the spirit and scope of the appended claims.

[0033] Existing accelerated degradation models based on step stress are mostly designed for products with a single performance index. However, many products have multiple performance indices, and these indices may all exhibit a certain degradation trend in degradation tests, requiring joint modeling of their degradation processes. To address this need, existing models use a combination of stochastic processes and copula functions, a method that is complex for engineering applications and difficult to select the optimal copula function. While multivariate degradation trajectory models, with their clearly defined variables and relatively simple operation, have not yet been widely applied to accelerated degradation modeling research based on step stress. Therefore, this application proposes a multivariate accelerated degradation modeling and reliability assessment method based on multivariate degradation trajectory models for accelerated degradation data of products with multiple performance indices. Please refer to [reference needed]. Figure 1 Specifically, it includes the following steps: S100: Obtain step stress accelerated degradation data and generate a dataset.

[0034] In this embodiment, when designing the accelerated degradation test using step stress, a total of [number] tests were set up. l There are several accelerating stress levels, where it is assumed that the following is adopted: This indicates the stress level of the product under normal operating conditions, while That is l A higher stress level.

[0035] During the experiment, the number of samples participating in the experiment was recorded as follows: n The number of performance indicators for each sample is k Each sample was placed sequentially in the above-mentioned... l In the acceleration stress level, at the _ ... q The number of degradation observations at each accelerated stress level is , q = 1,2, …, l Assume the sampling time of the degraded data is... ,in This indicates the first observation time, while , ,…, These represent the last observation time at each stress level. Therefore, the stress level at different observation times can be denoted as:

[0036] Assume the first i ( i = 1, 2, …, n The measured degradation values ​​of each sample at each observation time are represented by the following vector: ,in ( ), , , …, Indicates the product k ( k ≥ 1) performance indicators, and specify .like Figure 2 The figure shows the relationship between stress level and observation time.

[0037] Thus, the step stress accelerated degradation dataset is obtained. .

[0038] S200, Establish a multivariate step stress accelerated degradation model.

[0039] In this embodiment, the observation time is taken as ( Taking (e.g.) as an example, the modeling process for the observed degradation is explained. Specifically, for the observation time (e.g.) The corresponding degradation observations are modeled as follows:

[0040] in, It's about stress level. A continuous, monotonic function, representing the first... i The first sample p Performance indicators at a given stress level Stress influence coefficient under the given conditions . The parameter is A continuous monotonic function, representing the th p Baseline cumulative damage for each performance metric. , i = 1, …, n express n There are three independent and identically distributed Gaussian noises with a mean of 0 and a standard deviation of . , indicating that in the fitting of the th pUnexplained fitting errors that arise during the degradation process of a performance metric.

[0041] In this embodiment, The model is a product of the following two parts:

[0042] in, The form can be based on different accelerated degradation models. In this embodiment, the accelerated degradation model can be implemented using any of the following models: Arrhenius model:

[0043] Inverse power law model:

[0044] Exponential model:

[0045] To further introduce the heterogeneity between samples and the correlation between the degradation processes of different performance indicators, in this embodiment, It can be expressed as a product in the following form:

[0046] in, It is a fixed parameter, and It is a random parameter. , ,…, The joint distribution is assumed to be a multivariate normal distribution as follows:

[0047] in, Represent a k A unit vector of dimension, and It is k order variance The covariance matrix. Therefore, we can obtain:

[0048] in, .make

[0049]

[0050]

[0051] At this point, for the first stress level The degradation data collected below, i.e. The multi-element step stress accelerated degradation model proposed in this invention can be expressed as:

[0052] Considering the continuous nature of the degradation process, the above model is further extended to all... l Under a stress level. Assuming the observation time... In the first u A stress level, that is ( ):

[0053] in, , It is the stress influence coefficient matrix, which contains all k The performance index at the first u ( u = 1, 2, …, l The stress influence coefficients at each stress level are: The first in p ( p = 1, 2, …, k ) diagonal elements Indicates the first p The performance index at the first u Acceleration stress level Stress influence coefficient under the following conditions Similarly; It is the baseline cumulative damage matrix, which contains all k Each performance metric has a cutoff time. Baseline cumulative damage at time, where, The first in p ( p = 1, 2, …, k ) diagonal elements Indicates the first p Each performance metric has a cutoff time. Baseline cumulative damage at time Similarly; It is a random effects vector, which contains all k The heterogeneity of the degradation trajectory of each performance index among different samples, where, for The first in p ( p = 1, 2, …, k ) elements, For fixed parameters, To obey k Random parameters of a dimensional normal distribution; It is a random error vector, representing the error at the observation time. The first to last fit kUnexplained fitting errors that arise during the degradation process of a performance metric.

[0054] S300. The dataset is fitted using the multivariate step stress accelerated degradation model to estimate the model parameters.

[0055] Based on the S200-based multivariate step stress-accelerated degradation model, there are (1) in,

[0056] , , (2) Step stress accelerated degradation dataset obtained based on S100 , For the first i The sample in the first j Observation time By using the measured values ​​of degradation to fit the above model, the model parameters can be estimated. Specifically, in formula (2), , i = 1, 2, …, n If treated as unknown fixed parameters, the parameters in the first term can be estimated using multivariate least squares. Define the sum of squared errors. SSE for: (3) At this time, you can SSE right Find the first derivative and set it to zero, then we get: estimator : (4) In formula (4) ,…, Substitute the algebraic expression into formula (3) and minimize SSE , can be obtained , The estimation result is denoted as Substitute into formula (4) again The value can be obtained ,…, The numerical results.

[0057] Finally, based on the maximum likelihood estimation method, we can obtain... and The estimators are denoted as follows: and Its expression is as follows: (5) The ". / " symbol indicates that the elements at corresponding positions in the two matrices are divided.

[0058] Fitting error in the model variance The maximum likelihood estimate of the covariance matrix is: (6) Among them, matrix The square root of the diagonal element is the value of , which is . , …, The estimate.

[0059] S400: Based on the estimated model parameters and the multivariate step stress accelerated degradation model, the product reliability of the target product under normal use conditions is calculated.

[0060] After completing the modeling and parameter estimation of the multivariate step stress accelerated degradation model, the product reliability can be assessed based on the actual stress level.

[0061] In this embodiment, the failure criterion for the product is defined as follows: if the degradation of any performance indicator exceeds a given value, the product is considered to have failed.

[0062] Based on the multivariate step stress-accelerated degradation model mentioned above, for the degradation of the product at stress level S0, its... p The degradation process of a performance metric can be represented as: (7) It should be noted that the variables in the above formula no longer have subscripts "". i This is because the focus here is on the product as a whole, not on a specific individual; and it no longer includes a subscript. j This is because at this point we are considering the actual continuous degradation process of the product rather than discrete observations.

[0063] Let the product's failure threshold be determined by a vector. , ( ) represents a value where each element represents the failure threshold of a performance metric. Given a random variable... The reliability of a product can be expressed as: (8) in, The cumulative distribution function represents the standard normal distribution. Random variable. It follows a multivariate normal distribution, and its probability density function is: (9) In formula (8), the random variable By integrating, we can obtain the overall reliability function of the product at stress level S0 as follows: (10) The high-dimensional integral in the above formula can be approximated using the Monte Carlo method.

[0064] This application presents a statistical model for data collected from typical accelerated degradation tests of products with multiple performance indicators. The model considers both the heterogeneity between samples and the correlation between performance indicators, and also incorporates the influence of accelerated stress level on the degradation process. It features a clear model form, simple parameter estimation, and ease of application.

[0065] To further verify the effectiveness of the reliability assessment method proposed in this application, the application of the proposed model will be illustrated through simulation experiments below.

[0066] Without loss of generality, suppose a product has k = 3 key performance indicators, and in the accelerated degradation test of step stress, there are a total of l = Three different temperature stress levels, namely 100℃, 110℃, and 120℃. Therefore, in the proposed model, the stress function can be defined based on the Arrhenius model as... ,in p = 1, 2, 3. The model parameters are assumed to be:

[0067] ,

[0068] Assume there is n = 10 samples participated in the test, and at each temperature stress level, the number of degradation data observations for these samples was 5, that is Based on the above settings, a simulation dataset was generated. Figure 3 The simulation degradation data for the first performance index of all samples are shown, along with their correspondence with the three temperature stress levels.

[0069] Based on the case study above, the reliability assessment process is as follows: First, the degradation data generated in this simulation case is collected to form a dataset.

[0070] Then, the collected degradation dataset was fitted using the multivariate step stress accelerated degradation model proposed by S200. In the model, the stress influence coefficient function was defined based on the Arrhenius model. ,p = 1,2, 3.

[0071] Specifically, given the collected degradation dataset, the proposed multivariate step stress-accelerated degradation model is used to fit it, and the estimated model parameters are shown below:

[0072] ,

[0073] Figure 4 Taking five randomly selected samples from ten simulation samples as examples, the fitting effect of the proposed model on the simulated degradation data is demonstrated. It can be seen that the fitted degradation trajectory (shown by the dashed line in the figure) fits well with the actual degradation trajectory (shown by the solid line in the figure), and it can be considered that the proposed model fits this set of simulated degradation data very well.

[0074] Finally, in practical engineering, extrapolating the reliability level of a product under normal operating stress levels based on accelerated degradation test data is a crucial task. As an example, the product's failure threshold is defined here as... The temperature stress level of a product under normal use conditions is defined as... S 0 = 90℃. Based on the model parameter estimation results and formula (10), the reliability of the product can be calculated, such as... Figure 5 As shown.

[0075] Based on the same technical concept, embodiments of this application also provide an electronic device that can implement the multi-step stress accelerated degradation modeling and reliability assessment method provided in the above embodiments of the present invention. In one embodiment, the electronic device can be a server, a terminal device, or other electronic equipment. Figure 6 As shown, the electronic device may include: At least one processor and a memory connected to the at least one processor. In this embodiment of the invention, the specific connection medium between the processor and the memory is not limited. Figure 6 The example used is the connection between the processor and memory via a bus. The bus... Figure 6 The connections between other components are indicated by thick lines and are for illustrative purposes only, not as limiting information. Buses can be divided into address buses, data buses, control buses, etc., but for ease of representation, [the specific bus type is not shown here]. Figure 6 The processor is represented by a single thick line, but this does not imply that there is only one bus or one type of bus. Alternatively, a processor can also be called a controller; there are no restrictions on the name.

[0076] In this embodiment of the invention, the memory stores instructions executable by at least one processor. By executing the instructions stored in the memory, the at least one processor can perform the multi-step stress accelerated degradation modeling and reliability assessment method described above. The processor can implement... Figure 6 The functions of each module in the device shown.

[0077] The processor is the control center of the device. It can connect to various parts of the control device through various interfaces and lines. By running or executing instructions stored in memory and calling data stored in memory, it can monitor the device's various functions and process data, thereby enabling overall monitoring of the device.

[0078] In an alternative design, the processor may include one or more processing units. The processor may integrate an application processor and a modem processor, wherein the application processor primarily handles the operating system, user interface, and applications, while the modem processor primarily handles wireless communication. It is understood that the modem processor may also not be integrated into the processor. In some embodiments, the processor and memory may be implemented on the same chip; in some embodiments, they may also be implemented separately on separate chips.

[0079] The processor can be a general-purpose processor, such as a CPU, digital signal processor, application-specific integrated circuit, field-programmable array, or other programmable logic device, discrete gate or transistor logic device, or discrete hardware component, capable of implementing or executing the methods, steps, and logic block diagrams disclosed in the embodiments of this invention. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the multi-step stress accelerated degradation modeling and reliability assessment method disclosed in the embodiments of this invention can be directly manifested as being executed by a hardware processor, or executed by a combination of hardware and software modules within the processor.

[0080] Memory, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules. Memory can include at least one type of storage medium, such as flash memory, hard disk, multimedia card, card-type memory, random access memory (RAM), static random access memory (SRAM), programmable read-only memory (PROM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), magnetic memory, magnetic disk, optical disk, etc. Memory is any other medium capable of carrying or storing desired program code in the form of instructions or data structures, and accessible by a computer, but is not limited thereto. In embodiments of the present invention, memory can also be a circuit or any other device capable of implementing storage functions, used to store program instructions and / or data.

[0081] By designing and programming the processor, the code corresponding to the multi-step stress accelerated degradation modeling and reliability assessment method described in the foregoing embodiments can be embedded into the chip, enabling the chip to execute the steps of the methods described in the foregoing embodiments during operation. How to design and program a processor is a technique well-known to those skilled in the art and will not be elaborated upon here.

[0082] Based on the same inventive concept, embodiments of the present invention also provide a storage medium storing computer instructions that, when executed on a computer, cause the computer to perform a multivariate step stress accelerated degradation modeling and reliability assessment method described above.

[0083] In some alternative embodiments, the present invention also provides that various aspects of the multivariate step stress accelerated degradation modeling and reliability assessment method can also be implemented in the form of a program product, which includes program code that, when the program product is run on a device, causes the control device to perform the steps in the multivariate step stress accelerated degradation modeling and reliability assessment method according to various exemplary embodiments of the present invention as described in this specification.

[0084] It should be noted that although several units or sub-units of the apparatus have been mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to embodiments of the invention, the features and functions of two or more units described above can be embodied in one unit. Conversely, the features and functions of one unit described above can be further divided and embodied by multiple units. Furthermore, although the operation of the method of the invention is described in a specific order in the drawings, this does not require or imply that these operations must be performed in that specific order, or that all the operations shown must be performed to achieve the desired result. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step, and / or one step may be broken down into multiple steps.

[0085] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0086] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a server, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0087] Program code for performing the operations of this invention can be written using any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computing device, partially on the user's device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0088] In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0089] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0090] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0091] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for multivariate step stress accelerated degradation modeling and reliability assessment, characterized in that, include: Acquire step stress accelerated degradation data of the target product and generate a dataset; Establish a multivariate step stress accelerated degradation model; The dataset was fitted using the aforementioned multivariate step stress accelerated degradation model to estimate the model parameters; The reliability of the target product under normal use conditions is calculated based on the estimated model parameters and the multivariate step stress accelerated degradation model.

2. The multi-step stress accelerated degradation modeling and reliability assessment method according to claim 1, characterized in that, The acquisition of accelerated degradation data by step stress and the generation of a dataset specifically include: The number of samples participating in the experiment was set as follows: n The number of performance indicators for each sample is k The number of accelerated stress levels is l In the q The number of degradation observations at each accelerated stress level is , q = 1, 2, …, l ; Each sample was placed sequentially l In each accelerated stress level, the accelerated stress level increases sequentially, and the observation time is divided for each accelerated stress level to obtain the measured value of degradation of each sample at each observation time. The measured value of degradation consists of the performance index of at least one product. The dataset consists of the observation time, accelerated stress level, and measured degradation value for each sample.

3. The multi-step stress accelerated degradation modeling and reliability assessment method according to claim 2, characterized in that, The establishment of the multivariate step stress accelerated degradation model specifically includes: Assuming the observation time In the first u Acceleration stress level Below, that is , u Values ​​range from 1 to l The degradation observations are modeled as follows: in, It is the stress influence coefficient matrix, which contains all k The performance index at the first u ( u = 1, 2, …, l The stress influence coefficients at each stress level are: The first in p ( p = 1, 2, …, k ) diagonal elements Indicates the first p The performance index at the first u Acceleration stress level Stress influence coefficient under the following conditions Similarly; It is the baseline cumulative damage matrix, which contains all k Each performance metric has a cutoff time. Baseline cumulative damage at time, where, The first in p ( p = 1, 2, …, k ) diagonal elements Indicates the first p Each performance metric has a cutoff time. Baseline cumulative damage at time Similarly; It is a random effects vector, which contains all k The heterogeneity of the degradation trajectory of each performance index among different samples, where, for The first in p ( p = 1, 2,…, k ) elements, For fixed parameters, To obey k Random parameters of a dimensional normal distribution; It is a random error vector, representing the error at the observation time. The first to last fit k Unexplained fitting errors that arise during the degradation process of a performance metric.

4. The multi-step stress accelerated degradation modeling and reliability assessment method according to claim 3, characterized in that, The accelerated degradation model can be any one of the Arrhenius model, the inverse power law model, and the exponential model.

5. The multi-step stress accelerated degradation modeling and reliability assessment method according to claim 3 or 4, characterized in that, The estimated model parameters specifically include: For the obtained dataset is , For the first i The sample in the first j Observation time The measured values ​​of degradation are used to represent the constructed multivariate step stress-accelerated degradation model as follows: in, , , , ; in, The parameters in the model are often estimated using the multivariate least squares method. The parameters are obtained using maximum likelihood estimation.

6. The multi-step stress accelerated degradation modeling and reliability assessment method according to claim 5, characterized in that, The The parameters in the model are commonly estimated using the multivariate least squares method, specifically including: Define the sum of squared errors (SSE) as follows: SSE pair Find the first derivative and set it to zero, then we get... estimator : estimate Substituting the algebraic expression into the SSE formula and minimizing the SSE, we obtain... The estimation results; Then redo Substituting into the SEE formula, we get Numerical results; based on The numerical results were obtained by using maximum likelihood estimation to obtain the parameters. and The estimate.

7. The multi-step stress accelerated degradation modeling and reliability assessment method according to claim 5, characterized in that, The The parameters in the model are obtained using maximum likelihood estimation, and specifically include: Fitting error variance The maximum likelihood estimate of the covariance matrix is matrix The square root of the diagonal element is used as The estimate.

8. The multi-step stress accelerated degradation modeling and reliability assessment method according to claim 5, characterized in that, The formula for calculating the product reliability is: in, This refers to the stress level of the target product under normal operating conditions.

9. An electronic device, characterized in that, include: At least one processor; and a memory communicatively connected to the at least one processor; The memory stores instructions executable by the at least one processor, which executes the instructions stored in the memory to perform the method as described in any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store instructions that, when executed, cause the method as described in any one of claims 1 to 8 to be implemented.

Images