HPPN-based Prognosis for Hybrid Systems

International audienceThis paper presents a model-based prognosis method for hybridsystems i.e. that have both discrete and continuous behaviors.The current state of the hybrid system is estimatedby a diagnosis process and the prognosis process uses thisstate estimation to predict the future states and to determinethe end of life (EOL) or the remaining useful life (RUL) ofthe system. The Hybrid Particle Petri Nets (HPPN) formalismis used to model the hybrid system behavior and degradation.A HPPN-based diagnoser has already been defined toprovide a current state estimation that takes uncertainty aboutthe system model and observations into account. We proposeto generate a prognoser from the HPPN model of the system.This prognoser is initialized and updated with the result of theHPPN-based diagnoser. It computes a distribution of beliefsover the future mode trajectories of the system and predictsthe system RUL/EOL. The prognosis methodology is demonstratedon a three tanks example