Set-membership Parameter Identification of Linear Systems with Multiplicative Uncertainties: A New Algorithm

This paper develops a new set-membership estimation algorithm for the identification of time-varying parameters in linear models, where both additive and multiplicative uncertainties are considered explicitly. We show that a recursive algorithm we develop is capable of providing nonconservative approximations to the feasible solution set. Examples are considered to demonstrate the utility and effectiveness of the algorithm, including an application to health monitoring in marine dual fuel engines.