Research On Reliabily Modeling Of Products Based On Accelerated Performance Degradation Data

The product reliability index is an important reference for product formulation,production,and maintenance.The main performance index of a product is often gradually degraded with time before failure,until the cumulative degradation of this performance exceeds the failure threshold of the product,and ultimately causes the product to fail.The process of product degradation includes a large amount of information that is reliable,accurate,and relevant to the product's life.Therefore,the use of accelerated degradation modeling techniques to evaluate product reliability information and life characteristics has become a research hotspot in the field of reliability engineering.The stochastic process model represented by the process fully considers the randomness of the performance degradation of the product in the normal working or experimental environment,and can describe the product's true degradation process,becoming an important technical means for accelerating the degradation modeling.This paper first studies the reliability degradation modeling method based on the Wiener process model.We discussed the effects of factors such as material and manufacturing process on the product,there are individual differences,and the performance is degraded.The Wiener process degeneration modeling method with random effects sets the Wiener process model parameters as a random variable.Describe the individual differences in the product.For the problem of how to set random parameters for the Wiener process model with random effects,a method based on sample separation estimation is used to provide the numerical and theoretical reference for setting the random parameters of the Wiener process.The reliability modeling methods of the Wiener degenerative process with random effects when setting different random parameters under accelerated degradation test scenarios are discussed.Then,this article takes a product with two related performance degradation indicators as the research object,uses the Copula function to describe the correlation between the two performance degradation indexes,and discusses the binary of the Wiener process based on the time scale transformation under normal working stress.Degenerate modeling methods.Further extended to the accelerated degradation of products,and taking into account the difference in degradation between products,a degradation modeling method based on a binary Wiener process with random effects.Finally,the non-parametric statistical method-local linear regression smoothing method was used to fit the average degradation curve of accelerated degradation product performance index.Using the relationship between relative acceleration factors and degradation time,the functional relationship between the accelerating stress and the product degradation data under the highest stress is established.The degradation data under each accelerating stress is equivalently converted to the highest stress,and then the relative accelerating factor is fitted.The stress relationship model estimates the life characteristics of products under normal stress,and further estimates the parameter values of different failure distributions based on these statistical characteristics,thereby evaluating the product reliability under different failure types.