A Model-Selected Approach for Life Prediction Based on Accelerated Degradation Life Test

Abstract

The life and stability of electronic products are important factors of safety and reliability of intelligent buildings. Accelerated degradation test is an important way to stimulate the potential defects of products, shorten the testing time, and obtain the life and performance of electronic products. For the selection of degradation model and life distribution model during accelerated degradation test, this paper adopts simulation method to analyse the difference of assessment results and its impact that may be caused by the inconsistency between fitting model and actual model, based on different degradation models and life distribution models; and put forward a life assessment method of accelerated degradation data based on model sorting, to reduce the risks and impact of wrong model selection on assessment results; and finally an example is given to illustrate the method.