Decentralized multi-loop auto-rolling controller for nonlinear large-scale roll-to-roll machines via velocity sensorless and tension loop adaptation approaches

This paper proposes a novel decentralized auto-rolling mechanism for nonlinear roll-to-roll printing machines, including loop adaptation tension regulators. The key contributions of this work to address the practical concerns pertaining to (a) model–plant mismatches, (b) decrease in the number of sensors, and (c) performance consistencies are summarized as follows: first, removal of web velocity loop dependency, improving auto-rolling control accuracy and lowering implementation costs without plant parameter value information [for concern (b)]; second, active damping velocity error stabilization with the closed-loop order reduction property and improved disturbance rejection capability using nonlinear disturbance observers [for concerns (a) and (c)];; and third, the online tension loop adaptation rule driven by the simple analytic form (transient performance improvement). The numerical simulation study verifies closed-loop effectiveness compared with the recent auto-rolling technique.