The objective of this paper is to design decentralised sliding mode controllers for fractional-order large-scale nonlinear systems. In the first step, a fully-decentralised fractional-order sliding mode controller with a novel integral sliding manifold is developed. Practical stability of the closed-loop system is fulfilled under the assumption that the interconnections among the subsystems are bounded with known upper bounds. However, in reality the uncertainties and interconnections upper bounds are unknown. Therefore in the next step, an adaptive-fuzzy structure is applied to approximate the interactions and uncertainties. Since the states of neighbour subsystems are considered as the fuzzy system inputs, this technique is known as semi-decentralised control strategy. Due to using the fractional integral sliding surface, the zero convergence of the sliding manifold has been analysed based on integer-order stability theorems. In addition, system tracking errors convergence is deduced from fractional-order linear stability theorems. Computer simulations present the performance of the suggested controllers in the presence of uncertainties and interconnections.

Keywords