Robust fault detection filtering for a class of uncertain nonlinear singular system

In this study, design of observer based Fault Detection (FD) is proposed for a class of Lipschitz singular systems, in which the nonlinearities appear both in the state and measured output equations. The system is supposed to have norm-bounded uncertainties in all the four realization matrices and the coefficients of disturbances. Synthesis of the H-/H∞ observer based Fault Detection is done through semi-definite programming and strict Linear Matrix Inequality (LMIs). The resulting H-/H∞ observer based fault detection guarantees asymptotic stability of the estimation error dynamics with pre-specified disturbance attenuation level. Furthermore, it is robust against parametric uncertainties.