Abstract
Disturbances due to the eccentric unwinding roll/roller can create significant tension variations in the web spans in multi-driven web transport systems. The drivers are interconnected by an elastic web material (a 14-micrometer thick Polypropylene film was used for this study). Specifically, eccentricity of an upstream roller such as an unwinding roll creates upstream tension variation, and the tension variation in upstream web spans is transported to downstream due to the interconnection between the web spans. However, it is not easy to obtain an accurate mathematical model of the web transport system, for the design of a tension controller, which contains several actuators, sensors, gears, bearing systems, and interconnected web spans.
In this paper, an adaptive fuzzy controller which automatically tunes its parameters in real-time is suggested to regulate tension variations in multi-spans due to an eccentric unwinding roll. In the proposed method, the input and output ranges are adjusted together according to continuous observation of tension error and change in the error. The proposed controller is implemented on a prototype web transport system which consists of three AC motors, real-time target controllers, unwinding and winding rolls with two-spans of web including sensors for the measurement of tensions and speeds of the web.
The Experimental results show that the proposed adaptive fuzzy controller successfully rejects the effects of the disturbances due to the eccentric unwinding roll and substantially reduces the tension variations in both web spans. The adaptive fuzzy controller outperformed the conventional PID controller as the frequency of the eccentricity disturbance increased.
