Reliability Evaluation Method of Performance Degradation Data Based on Proportional Risk Model under Small Sample Conditions

Abstract

With high reliability, high cost and multiple degradation characteristics of the product, carryout under the condition of small sample statistical analysis of degradation data has morepractical application value for using the degradation data to assess factors affecting thereliability of more complex or unknown products, given the proportional hazards model can beused to describe the product reliability level and the quantitative relationship betweenenvironmental factors, this paper presents a degradation failure rate based on proportionalhazards model analysis method to speed up the first time as a covariate factor, proportionalhazards model is introduced to estimate the product life and the relationship between thequantity of degradation failure rate, failure rate percentage degradation model is set up; Then,according to the relationship between reliability and the distribution of degradation quantity,the distribution fitting method is used to estimate the reliability indexes of products, so as toobtain the estimated value of unknown parameters, and then the reliability evaluation ofproducts is carried out. With laser emitter GaAs degradation data for example, make full use ofevery moment of the current increment value, calculate the moment of specific degradationfailure rate, the results show that the method can accurately describe the degradation process,is based on the traditional reliability method of failure distribution reflects the same averagecharacteristics of subjects environment with the type of product, and cannot be directly appliedto real-time quantitative reliability analysis, this paper used the method compared with othermethod has strong adaptability and accuracy, better able to complete the task of degradationdata reliability estimates.