Mode Tracking Techniques

Abstract

Monitoring of hybrid system requires measurement or estimation of continuous state-variables and tracking the system discrete dynamics (i.e., the system mode evolution). Some mode-changes are known, because they are initiated by a supervisory controller or triggered by measured continuous states. The main difficulty when applying model based monitoring techniques to hybrid systems is due to unpredicted mode-changes caused by unknown discrete inputs, and unknown discrete dynamics; these mode changes can happen at any time and in any order (in particular, if some of these modes represent faults). In this chapter, a mode tracking technique is introduced. To make the mode tracking more efficient, a unique mode-change isolation process is utilized, this process is based on the Mode Change Signature Matrix (MCSM). The mode-tracking is efficiently integrated into the health monitoring process and the mode tracker is invoked only when inconsistency between the monitored system and the model is detected by the FDI module. After that, a new method utilizing the partially known dynamical model to identify hybrid system modes in the presence of a single parametric fault is introduced. Mode identification of hybrid system in the presence of faults is based on continuous evaluation of all ARRs given by the ARR-set. Finally, mode identification and multiple fault estimation with unknown fault pattern is discussed.