Prognostic and health management for avionics

Abstract

This paper describes methodologies and tools to anticipate in real-time, the onset of failures in electronic equipment. Currently developed techniques for prognostics and health management (PHM), depend on the observation of precursor variables and an imputation from them of impending failure. In electronic systems, such variables are difficult, impossible or expensive to obtain, so we propose a model-based technique utilizing embedded life models and environmental information obtained from aircraft mounted sensors that measure those parameters that most significantly accelerate failure in electronics, i.e. temperature, temperature variation, vibration and shock. This technology is referred to throughout this paper as electronics prognostics and health management (ePHM).