ROBUST FAULT ISOLATION FILTER FOR NETWORKED CONTROL SYSTEMS: MIMO CASE WITH MULTIPLE FAULTS

Abstract The effect of network-induced delays introduced into the control loop is modelled as time-varying disturbance. A fault isolation filter (FIF) for fault detection of multiple input multiple output (MIMO) networked control systems (NCSs) with multiple faults is then parameterized. Directional residual generations decoupling from the disturbance ensure the treatment of multiple faults appearing simultaneously or sequentially. The remaining degrees of freedom in the design of the filter's gain are used to satisfy an H ∞ disturbance attenuation and poles assignment constraints in the frame of Markov jump linear systems (MJLS). The sufficient existence conditions of the free parameters are formulated as a convex optimization problem over a couple of linear matrix inequalities (LMIs). An illustrative example is given to show the efficiency of the proposed method for NCS.