A stochastic process based reliability prediction method for LED driver

Abstract

In this study, we present a general methodology that combines the reliability theory with physics of failure for reliability prediction of an LED driver. More specifically, an integrated LED lamp, which includes an LED light source with statistical distribution of luminous flux, and a driver with a few critical components, is considered. The Wiener process is introduced to describe the randomness of lumen depreciation. The driver's survival probability is described using a general Markov Chain method. The system compact thermal model (physics of failure model) is developed to couple with the reliability methods used. Two scenarios are studied: Scenario S1 considers constant driver's operation temperature, while Scenario S2 considers driver's temperature rise due to lumen depreciation. It has been found that the wide life distribution of LEDs will lead to a large range of the driver's survival probability. The proposed analysis provides a general approach for an electronic system to integrate the reliability method with physics models.