ELECTROMAGNETIC SUSPENSION - NEW RESULTS USING NEURAL NETWORKS

Control of magnetically suspended vehicles is an active area of current research. Due to nonlinear magnet characteristics, controllers for these systems are usually designed by using linear small perturbation models. This assumption of linearity poses serious constraints on the operating range (allowable airgap) of the suspended objects. This paper presents a brief account of using the neural network concepts in the formulation of the nonlinear airgap-force characteristics of an experimental electromagnetic suspension system. The main advantage of using the NN is the designers ability to identify the magnet-force characteristics through experimental data (learning).