Real Time Application of an AMB Using MLP: Study of Robustness

Abstract

The active magnetic bearing AMB presents a solution for all 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 control of the AMB by classical controllers. The study of robustness and the application of external disturbances, for a system controlled with MLP, were done in order to highlight the advantage of controlling this non linear system by neural networks. The study of the internal disturbance (the balourd) created while rotation is also applied in order to reduce the effect of this unbalance.Keywords-component: Non linear dynamics, adaptive control dynamics, adaptive learning network, artificial neural network applications, back propagation algorithm learning, and dynamic neural network control.

Meta-learning to optimize the number of hidden nodes of MLP networks trained by Extreme Learning Machine algorithm

“…ANNs have been successfully applied to different problems in the literature [12] [13] [14] [15]. The success of these algorithms in applications motivated the development of new models with different levels of complexity.…”