Adaptive Extremum Control by the Parametric Volterra Model

Abstract

The paper presents a strategy based on the parametric Volterra model. This model is linear in the parameters and has, nevertheless, an infinite memory. This approximation represents more exactly a general nonlinear dynamic model described by the Volterra series, than the Hammerstein model, having already been investigaged in several earlier papers. On the basis of this model, an adaptive extremura control algorithm is formulated according to the dual control theory. In order to ensure a rapid convergence to the true extremum value, an external perturbation signal was added to the control input.