Chaos Behavior in Nonlinear Servo System and Its Control

Abstract

The induction motor is superior to a dc motor in maintenance. Though the control of induction motor is much more difficult than that of dc motor, it is currently used by realization of so-called vector control method. But, this method doesn't establish the stability of the system, neither is robustness. The dynamics of the induction motor is represented by nonlinear equations with three-inputs and three-outputs. Though precise parameters are necessary for the vector control method, it is generally difficult to obtain the precise values. Therefore, there are often identified errors in parameters. In this case, as the effect of the nonlinear terms remains, it is known that chaotic motions may occur in the system. As chaotic motion isn't necessarily desired for the system, it's necessary to remove the chaotic behavior when this behavior occurs. In this paper, the existence of chaotic motion in the system can be verified, and the generating conditions are found using Lyapunov exponents. Next, a chaos control method using neural networks is proposed in order to remove the chaotic behavior.