Detection, identification, and compensation of nonlinearities and an experimental verification platform for nonlinear controllers.

摘要

Identification of nonlinear systems has been an active research area for more than one decade. But compensation of nonlinearity has not received much attention. This thesis investigates the problem of nonlinearity compensation in closed-loop systems from a holistic point of view. First, the effect of nonlinear distortions on closed-loop performance of systems is presented. Next, new nonlinearity models for identification of Wiener and Hammerstein systems are presented with the objective of canceling the effect of nonlinear distortions. Use of hyperbolic splines toward the compensation of static sensor nonlinearities is also explained.; This thesis also investigates the need for a platform for experimental verification of nonlinear control algorithms. Development of a fractional horsepower dynamometer is described for this purpose and experimental results are shown for five different nonlinear control schemes, namely, state-periodic adaptive control, adaptive friction compensation, adaptive feedforward cancelation, repetitive control, and sliding mode control.