Hybrid Health-Aware Supervisory Control Framework with a Prognostic Decision-Making
Résumé
This paper deals with the design of a Hybrid Health-Aware Supervisory Control (HHASC) scheme for Linear Time Invariant (LTI) systems. The proposed HHASC setup integrates a switching fault-tolerant controller, a task of Fault Detection and Isolation (FDI), and the use of prognostic information. This approach is able to manage current and future health situations to extend the system life by the consideration of control objective mitigations to reduce the load on degraded components. In addition, the introduction of the Prognostic Decision-Making (PDM) unit fitted with a Virtual Fault Mechanism (VFM) allows to forecast the switching FTC and reduce the reconfiguration transients. The dwell-time property and its dual part are used to derive an input/output closed-loop stability proof. The proposed HHASC framework provides then a cohesive setup where prognostic, diagnosis and accommodation tasks can work in harmony together. The proposed solution is finally applied on a numerical aeronautical benchmark to highlight its benefit.