Nonlinear Control of a Double-Electromagnet Suspension System A. Maghsoudlu 1 , R. Barzamini 2 , K. D. Farahani 2 1. Islamic Azad University- Aliabad Katul Branch 2. Iran Power Develeptment Company (IPDC) afshin_m_ir@yahoo.com, barzamini@yahoo.com, engfarahani@yahoo.com Abstract—In this paper a feedback linearization is presented that addresses the coupling effects between two groups of electromagnetic trains. The module, based on some reasonable assumptions of nonlinear mathematical model, is modeled as a double-electromagnet system. The proposed algorithm in tracking has a satisfying performance in presence of unknown changes in the mass. It also shows robustness against the presence of measurement noises which is the case in the considered plant where all sensors collect noise from the environment. The simulation results show the capability of the proposed algorithm in the presence of input and output perturbation. Keywords-component; Double-electromagnet, Suspension, Nonlinear, Feedback Linearization I. INTRODUCTION Today, express trains have significantly eased traffic congestions in urban areas. The basic idea for designing such trains is the magnetic suspension. As shown in Fig. 1, in a single suspension system, the mass is under the influence of two forces: the gravity and the magnetic force. In the case of passenger trains, a different type of suspension system is used (Fig. 2). In fact, the two groups of electromagnets embedded in a module are connected to a rigid body. Therefore their motion states are coupled. Currently, the general method controls the two groups of electromagnets separately using two independent controllers, each of them acts according to respective controlled object. The coupling between the two groups of electromagnets is regarded as disturbance and suppressed by enhancing the robustness of individual controllers. However, this method cannot actively overcome the uncertainty issues, and the control performance is not desirable especially in the presence of external disturbances. In recent years so many studies have been carried out on electromagnet suspension systems. Chen and Wang (2003) presented a larger moving range dual-axis magnetic-levitation (maglev) system. A repulsive maglev system with four active guiding tracks was adopted. The system was treated as a multi-input multi-output system. Wai (2008) developed the dynamic model of a maglev transportation system based on the concepts of mechanical geometry and motion dynamics, including levitated electromagnets and a propulsive linear induction motor (LIM). A model-based sliding mode control (SMC) strategy was introduced. Bonivento et al. (2005) addressed the problem of positioning a ball in a vertical magnetic field created by a pair of electromagnets while rejecting some external disturbances. The presence of uncertainties on the physical characteristic parameters of the system has been taken into account by designing a control law, capable of solving the problem robustly. De- Queiroz and Pradhananga (2007) presented a general formulation for constructing stabilizing control laws with a time-varying bias flux. They suggested a bias flux function for the mechanical states. General conditions were established on the bias function to ensure a singularity-free controller and power losses that converge to zero as the mechanical states converge to zero, without affecting the system stability. Peterson et al. (2006) showed experimentally that, by selecting the CLF based on the solution to an algebraic Riccati equation, it is possible to tune the performance of the controller using intuition from classical LQR control. Lin et al. (2005) proposed a hybrid controller using a recurrent neural network (RNN) to control a levitated object in a magnetic levitation system. A nonlinear dynamic model of the system is Fig. 1: Magnetic suspension system 978-1-4244-9093-6/10/$26.00 ©2010 IEEE 978-1-4244-9094-3/10/$26.00 ©2010 IEEE