Tension Control in Unwinding System Based on Nonlinear Parameter Estimation

Abstract

In this paper, two representative nonlinearities in the unwinding system are modelled. One is the rotation-disturbance caused by the rotational change of unwinding roll position; another is the time-varying radius of unwinding roll. Based on the nonlinear mechanism model, the parameters of rotation-disturbance are obtained by the Gauss-Newton algorithm, and the time-varying radius is estimated online by the Kalman filter. Then the feedforward and time-varying feedback controller are designed based on the estimated parameters to compensate for nonlinear effect in unwinding system. The experiment results show that the parameter estimation algorithm can accurately estimate parameters. And the use of estimated parameters to control can effectively solve the problem of tension fluctuation and tension deviation which caused by the nonlinearity of unwinding system.