Iterative Learning Velocity and Tension Control for Single Span Axially Moving Materials

Abstract

Precise tension and speed control of axially moving material systems enables high speed processing of paper, plastics, fibers, and films. A single span model is developed that includes distributed longitudinal vibration, a torque-controlled roller at the left boundary, and a speed-controlled roller at the right boundary. The speed trajectory of the right roller is assumed periodic but unknown. A proportional and derivative (PD) feedback and iterative learning control (ILC) feedforward control law is developed for the left roller torque based on the measured tension and speed at the left boundary. PD tension/speed control is proven to ensure boundedness of distributed displacement and tension. ILC is proven to provide the same theoretical result but greatly improved simulated response to an aggressive stop/start right roller speed trajectory.