Methods of virtual sensors design for nonlinear systems

Abstract

The problem of virtual sensor design for systems described by nonlinear models to solve the problems of fault diagnosis and control is proposed. Virtual sensors are designed based on Luenberger observers. The problem is solved in three steps: in the first step, the linear model invariant with respect to the disturbance is designed; in the second step, a possibility to take into account the nonlinear term and to estimate the given variable is checked; and finally, stability of the observer is obtained. The relations allowing to design a virtual sensor with minimal dimension estimating the given component of the state vector of the system are obtained. The method to ensure stability of the observer is analyzed.