Boundary control of an axially moving string via fuzzy sliding-mode control and fuzzy neural network methods

The objective of this study is to develop intelligent control schemes for the transverse vibration reduction of an axially moving string. The proposed approaches are backboned by the methods of fuzzy sliding-mode control (FSMC) and fuzzy neural network control (FNNC). In practice, the control effort for the system is realized through a typical mass–damper–spring (MDS) system attached at the right-hand side boundary of the moving string. The dynamic coupling between the string and the MDS system provides an actuation force to suppress transverse vibration. In the first phase of this study, the framework of FSMC is designed, in which the techniques of region-wise linear fuzzy logic control design and generic algorithm technique are employed to facilitate FSMC to reduce a large number of fuzzy rule bases and to select optimal control gains, respectively. In the second phase, the FNNC is developed, which is, compared to the FSMC, easier to design the control rule, more robust against environment and capable of on-line learning. Numerical simulations are conducted and the comparison between various controllers is made based on simulations. The simulated results show that the transverse vibration can be well suppressed by both approaches. FSMC offers the capability to regulate the transient response, while FNNC holds advantage of on-line learning capability.