Development of Magnetic Levitation System for Science and Technology Education: Magnetic Levitation Control by Using a Hall Element Displacement Sensor with Neural Network

Abstract

In this study, we developed a magnetic levitation system using a Hall element displacement sensor with neural network for science and technology education. The sensor configured with three Hall elements was devised in order to measure displacement from an electromagnet to a levitated object with a permanent magnet. Use of the Hall element displacement sensor achieves a lower-cost magnetic levitation system. Furthermore, three-layered feedforward neural network was utilized in order to improve the precision of the Hall element displacement sensor. Finally, operation verification of the developed magnetic levitation system was conducted by designing state feedback regulator with observer.