Search

Many new techniques have been developed in recent years. One of the most notable techniques is the magnetic levitation train, or Maglev. A Maglev uses electro-magnetic control systems to levitate a vehicle in a short distance away from a guide way vertically. The idea was firstly patented in Germany by 'Transrapid'. After a few decades of development, the Maglev train has been already used for public service in China. The project studied a similar dynamic control system to control the lift of a maglev train. An ECP Mode 730 Magnetic levitation plant was used in the development of the control system. The system modelling was identified first and then a PID controller were designed, simulated and implemented. It was found that the characteristics of a PID controller are not good enough to such a maglev plant which requires a quicker response and almost no overshoot. A deadbeat controller later was designed to handle the maglev system which could give a much quicker response and no overshoot. The simulation results for a deadbeat controller suggested the overshoot of the system when the step input is 2cm is 0.013 (less than 0.0065%) which can be almost neglected. The settling time for the system response is 0.231 seconds and it has only 0.031 seconds‟ difference from the desired time. While applying the designed deadbeat controller to the plant, some real-world problems such as oscillations and control errors occurred. The problem was solved at the end and system performance became much better but the small oscillation still exists. It was believed that the small oscillation was coming from the hardware of maglev plant itself. In comparison with a classic PID controller, it was found the settling time has been improved at least 55% at the linearization point and the overshoot was reduced. However, when it comes up to a large movement from the linearization point there was no improved at all. There is a need to apply adaptive control techniques in the further work.