A Direct-Decoupling Closed-Loop Control Method for Roll-to-Roll Web Printing Systems

The roll-to-roll (R2R) web printing system is a complex coupling system, in which tension fluctuation is caused by upstream register control, thus results in downstream register errors. Therefore, it is indispensable to compensate for the couplings in order to improve the register precision in R2R printing systems. However, existing control methods do not realize complete decoupling due to their indirect calculation of compensations for the register errors. In this article, a mechanical model is set up to represent the direct relationship between the downstream register errors and all their upstream register controls. According to the model, compensation is calculated on the basis of the Lyapunov stability theorem to converge the register errors to zero. Then, a direct-decoupling closed-loop control method with first-order compensation terms, i.e., the direct-decoupling proportional derivative control (DDPD), is proposed to completely compensate for the couplings between all upstream register controls and downstream register errors. In addition, the first-order expression of the compensation makes it easy to implement in industrial applications. An industrial example indicates that the proposed control method eliminates the couplings and maintains the range of register errors within ±0.06 mm.
Note to Practitioners
—This article proposes a control strategy for the roll-to-roll (R2R) printing system, particularly for printing systems with electronic line shafts. Few existing studies have been done in detailedly analyzing the upstream register controls and the downstream register errors. This article establishes a mechanical model of printing registration and gives a systematic analysis of the complete relationship between upstream register controls and downstream register errors and then proposes a control strategy based on the Lyapunov stability analysis. The proposed method can be extended to other similar R2R systems. Simulation and industrial examples show that the control method is more feasible and effective compared with the existing methods. In future work, we will study a control method combined with the mechanical model and data model in different R2R systems.