Modeling of belt conveyor system based on least square support vector machine

Abstract

The paper presents nonlinear modeling study of belt conveyor system using the least square support vector machine (LS-SVM). Belt conveyor is a nonlinear, severe disturbance and time-varying system. So far, most of the existing models are based on mechanism laws, which are very useful for belt conveyor design. However, they are too complicated to be applied to control system design. To facilitate a valid control strategy design, this paper tries to avoid the internal complexities and presents a black-box model of the belt conveyor based on LS-SVM. The simulation tests reveal that it is feasible to establish the model using LS-SVM. Compared with the model based on the NARMA model, the simulation results are shown that the model based on LS-SVM algorithm is superior in data fitting.