Nonlinear Control of a Magnetic Levitation System Using Single Multiplicative Neuron Models

Abstract

The paper presents an approach to nonlinear control of a magnetic levitation system using artificial neural networks (ANN). A novel form of ANN, namely, single multiplicative neuron (SMN) model is proposed in place of more traditional multi-layer perceptron (MLP). SMN derives its inspiration from the single neuron computation model in neuroscience. SMN model is trained off-line, to estimate the network weights and biases, using a population based stochastic optimization technique, namely, particle swarm optimization (PSO). Both off-line training and on-line learning of SMN have been considered. The ANN based techniques have been compared with a feedback linearization approach. The development of the control algorithms is illustrated through the hardware-in-the-loop (HIL) implementation of magnetic levitation in LabVIEW environment. The controllers based on ANN performed quite well and better than the one based on feedback linearization. However, the SMN structure was much simpler than the MLP for similar performance. The simple structure and faster computation of SMN have the potential to make it a preferred candidate for implementation of real-life complex magnetic levitation systems.