A control system design of automatically tuning controller for roll to roll web system using the modified genetic algorithm

In the last decades, there are many applications which employed roll-to-roll web technology for mass production such as web printing, papers machine, film processing, and textiles fabrics and so on to make cheaper production in shorter time. Especially, RFID and printed electronics using the principle of roll-to-roll manufacturing to build equipment at high speeds will have a big impact on the printed electronics and publishing industries. In order to increase the precision and flexible ability to detect effectively the external and internal disturbance, several controller designs for roll-to-roll web control system are proposed in recent years. However, almost proposals are employed for the time-invariant linear SISO systems. In reality, some terms in several control systems such as roll-to-roll web system change in time and are non-linear. Therefore, classical gain tuning approaches are not suitable. In this paper, an algorithm proposed for non-linear roll-to-roll web control system is based on using the modified genetic algorithm to determine the optimal parameters of controller, detect the errors and automatically recover the errors at every time interval. The simulation results in Matlab/Simulink employed firstly with the simple plants and after that with back-stepping controllers of roll-to-roll system prove the stability and reliability of the proposed approach.