Adaptive feedforward control of Hammerstein non-linear systems

Abstract

Adaptive deadbeat feedforward control for measurable disturbances is analysed for a single-input, single-output non-linear system represented by the Hammerstein model wherein the non-linearity is any odd order polynomial. The recursive least squares estimation method of Chang and Luus (1971) is employed for parameter estimation. Simulation results show that the non-linear adaptive feedforward controller is superior to that of the linear adaptive feedforward controller.