Nonlinear Friction Compensation Method in Linear Motors Using Neural Network

Abstract

In this paper, we propose a method to compensate for nonlinear friction in linear dc motors (LDM) using a neural network controller that is added to the conventional controller. The output of the neural network controller is multiplied by sign function of the velocity reference, because it learns rapidly the nonlinear friction which is related to signs of the velocity. We applied the new method to positioning control of an LDM. The validity of the new method was verified by the experimental results.