A new approach of modelling power systems for robust control

This paper presents a power system model transformation, using the mean-value theorem, to an equivalent linear model which is made up of two Jacobian matrices - one evaluated at the equilibrium point and another at the mean-value point. The Jacobian at the equilibrium point is normally used for control design and the other Jacobian in considered as unmodelled dynamics or uncertainty. In this paper it is shown that a tighter bound on the unmodelled term is possible using this transformed model. This paper further presents the use of a genetic algorithm to calculate this bound.