Identification and control of a non-minimum phase flexible dynamical system using neural networks

In this paper, a two-degree-of freedom control system for motion control of a flexible system is presented. The two-degree-of-freedom control system consists of a feedback controller and feedforward controller. The feedback controller is designed based on Lyapunov´s direct method while the feedforward controller is designed based on the inverse dynamics. Several computational approaches are compared for making an inverse dynamics model of a control object. Three different approaches are shown to acquire the inverse dynamics model: a mathematical model based on the Laplace transformation method; a system identification model based on ARX model; and a neural network identification model. Experimental and numerical results for the tracking control of a self-sensing piezo-polymer actuator are presented and verify that the proposed control system is effective for controlling flexible dynamical systems