Neural networks backstepping sliding mode control approach based fractional-order calculation for Roll-to-Roll systems

In this paper, an adaptive controller is designed for three-shaft roll-to-roll systems with uncertainty and disturbance elements. By utilizing backstepping technique, sliding mode surface theory, and fractional-order calculation, we propose a novel control structure for driving three motors of the system. Meanwhile, two neural networks are applied to avoid issues like eccentric rollers and variations of radii that continuously affect the system’s ability to perform during operation. To demonstrate the correctness of the proposed controller, we consider the stability of the entire system as well as the control performance via comparative simulation. The obtained results indicate that the control strategy remarkably ameliorates the control ability of output tension and velocity.