Backstepping-Sliding Mode Control Combined with Load Torque Neural Network Observer for a Two – Mass System

Abstract

This paper describes the design and the simulation of a non-linear controller for a two-mass system (TMS) based on the backstepping-sliding mode control combined with a load torque neural network observer. The aim is to control the actual angular speed matching with the angular reference speed. The backstepping-sliding mode controller is designed based on the Lyapunov standard. The proposed neural network can torque estimate to approximate an appropriate value for unknown factors even when it is affected by the bounded disturbance.