|
|
 |
| |
|
|
|
Nonlinear Autoregressive Neural Network with Exogenous Input Model Approach for Magnetic Flux Density Measured by Hall-Effect Sensor in Magnetic Spring |
|
|
|
PP: 87-99 |
|
|
|
doi:10.18576/amis/190108
|
|
|
|
Author(s) |
|
|
|
Grazia Lo Sciuto,
Joanna Bijak,
Zygmunt Kowalik,
Tomasz Trawin ́ski,
Nadir Omer,
Mohammed Sallah,
|
|
|
|
Abstract |
|
|
| In this study, the magnetic and mechanical properties of the magnetic spring system are investigated. Experimental tests of the levitating magnet displacement in the magnetic spring are conducted by the laser distance meters and the detection of the magnetic flux density in the magnetic spring is provided by the three Hall-effect sensors. The measurements of the displacement and magnetic flux density in the magnetic spring excited by the vibration generator are characterized by nonlinear behaviour. The nonlinear mathematical model is proposed to predict and approximate the magnetic flux density starting from geometrical properties, voltage of vibration generator, frequency and displacement of the levitating magnet based on the nonlinear autoregressive networks with exogenous input (NARX) neural network architecture. The accuracy of the results obtained by NARX emphasises as the modeling technique can be used for construction and design of non-linear magnetic spring devices.
|
|
|
|
|
|
|
|
