A new method of controlling active magnetic bearing through neural network

The active magnetic bearing AMB presents a solution for all the technical problems since it ensures the total levitation of a body in space eliminating any mechanical contact between the rotor and the stator. The goal of our work is to show that the control of the AMB by Multilayer perceptrons MLP involves an improvement of the response compared to the ordering of the AMB by classical controllers. Our team has developed several diagrams with MLP to control the AMB. A final diagram was used and in which we optimized all the parameters influencing the training in order to obtain better results concerning the temporal answers of the positions of the axes.