Controller design for a nonlinear servomechanism in the presence of system uncertainties using multilayered neural network

Abstract

The problem of designing a nonlinear proportional plus integral controller to solve the nonlinear servomechanism problem is considered. Specifically, it is desired to find a controller for a plant to solve the servomechanism problem in the presence of nonlinear system uncertainties, so that the closed loop stability and asymptotic regulation occurs. A new optimal controller has been developed for a nonlinear dynamic system in the presence of system uncertainties with quadratic cost function. It is made up of two controllers: proportional plus integral feedback controller (PI) is designed using a nominal linear system model to eliminate the steady state error due to step input function, and the multilayered neuro controller using back propagation error algorithm is used to add nonlinear effect on the nominal system to compensate for nonlinear system uncertainties that are not considered in the classical linear optimal regulator (LOR) design of the nominal system. Computer simulation is performed to show the applicability and limitations of the proposed controller.