Neural network based intelligent control and PID control of a magnetic levitation system
In order to control a system effectively, the system must be modeled accurately. Frequently, the non-linearities present in system dynamics make the control task difficult. Sometimes it is a real challenging task to come up with a true dynamic model of the system. To appreciate the control complexities, a Magnetic Levitation (ML) System is selected for laboratory demonstration purposes. The magnetic levitation system, wherein the primary objective is to balance a metallic ball in a magnetic field, is highly non-linear by its very nature. For this paper, a dynamic model was derived for such a system and various control techniques were designed and applied and the system performance was compared. A neural network based controller was developed to control the system. Such controllers are particularly useful in those cases, where the mathematical model of the system may not be available. Proportional (P), and Proportional plus Integral plus Derivative (PID) controllers were the other controllers used for the study, and their performances were compared with the neural net work based controller.
American Society of Mechanical Engineers, Dynamic Systems and Control Division (Publication) DSC
Volume / Issue
Start / End Page