eduzhai > Applied Sciences > Engineering >

Modeling of Maglev System

  • Save

... pages left unread,continue reading

Document pages: 9 pages

Abstract: This paper proposes a simple nonlinear modeling technique using least squares support vector machines (LS-SVM) which approximates nonlinear dynamics by repeated training and validation using crossvalidation or leaveoneout method efficiently. A Radial Basis Function (RBF) kernel is applied for nonlinear system prediction in which the kernel parameters are tuned by Coupled Simulated Annealing (CSA) and simplex method to obtain accurate prediction to implement nonlinear predictive control. The proposed method matches the measured data more accurately than the feed forward artificial neural network (ANN) model. A comparative study on a Magnetic Levitation (Maglev) system proves its better prediction capability than Neural Network model.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×