Fault detection and fault tolerant control of a smart base isolation system withmagneto-rheological damper

Abstract

Fault detection and isolation (FDI) in real-time systems can provide early warnings forfaulty sensors and actuator signals to prevent events that lead to catastrophic failures. Themain objective of this paper is to develop FDI and fault tolerant control techniques for baseisolation systems with magneto-rheological (MR) dampers. Thus, this paper presents afixed-order FDI filter design procedure based on linear matrix inequalities (LMI). Thenecessary and sufficient conditions for the existence of a solution for detecting and isolatingfaults using the formulation is provided in the proposed filter design. Furthermore, an FDI-filter-basedfuzzy fault tolerant controller (FFTC) for a base isolation structure model was designed topreserve the pre-specified performance of the system in the presence of various unknownfaults. Simulation and experimental results demonstrated that the designed filter cansuccessfully detect and isolate faults from displacement sensors and accelerometers whilemaintaining excellent performance of the base isolation technology under faulty conditions.