Abstract—In this paper an intelligent method is proposed for controlling a kind of magnetic suspension system with 3 degree of freedom. At first, the dynamic of the magnetic suspension system and the related equations are presented. Regarding unstable nature and non-linearity of magnetic suspension system using techniques of liner control for achieving optimal performance so that all requirements of system are met in all domains is difficult. Then optimal controlling input for magnetic suspension system is designed using optimal control method, linear quadratic regulator (LQR) and required computations. For designing the neural network controller, Radial Basis Function (RBF), we use the results gained by LQR controller. The simulation results are performed using MATLAB software and performance of proposed controlling method was approved.

Index Terms—Magnetic suspension, Linear quadratic regulator, Radial basis function, Neural controller.

M. Saberi and S. M. Alizadeh are with the Faculty of Electrical and Computer Engineering, Lahidjan Islamic Azad University, Lahidjan, Iran (e-mail: saberi933@yahoo.com ; s.morteza.alizadeh@gmail.com).
H. Altafi is with the Sama Lahidjan College, Lahidjan, Iran and Gilan Zolal Company, Rasht, Iran (e-mail: hamid.altafi@gmail.com).

Cite: Mohammad Saberi, Hamid Altafi, and Seyyed Morteza Alizadeh, "Control of the Magnetic Suspension System with a Three-degree-of-freedom Using RBF Neural Network Controller," International Journal of Computer and Electrical Engineering vol. 4, no. 2, pp. 121-126, 2012.