Experiments with Neural Networks in the Identification and Control of a Magnetic Levitation System Using a Low-Cost Platform

Abstract

In this article, we designed and implemented neural controllers to control a nonlinear and unstable magnetic levitation system composed of an electromagnet and a magnetic disk. The objective was to evaluate the implementation and performance of neural control algorithms in a low-cost hardware. In a first phase, we designed two classical controllers with the objective to provide the training data for the neural controllers. After, we identified several neural models of the levitation system using Nonlinear AutoRegressive eXogenous (NARX)-type neural networks that were used to emulate the forward dynamics of the system. Finally, we designed and implemented three neural control structures: the inverse controller, the internal model controller, and the model reference controller for the control of the levitation system. The neural controllers were tested on a low-cost Arduino control platform through MATLAB/Simulink. The experimental results proved the good performance of the neural controllers.

Recurrent neural network based high-precision position compensation control of magnetic levitation system

“…Finally, a nonlinear autoregressive exogenous-type neural network controller (NARX-NNC) 22 is utilized to compare with the proposed controller. For the sinusoidal signal tracking, the RMSE using NNC is 0.0472, which is obviously larger than the 0.0145 obtained by the proposed controller.…”

“…Then, based on Kirchoff’s law and Biot-Savart’s law, the electromagnetic force can be calculated as follows 22 where , and parameters , , and represent the vacuum permeability, permeability area, and coil turn respectively.…”

“…For example, Rubio et al presented a neural network controller consisting of a nonlinear method and a neural network and reduced the root mean square error of the trajectory tracking in the magnetic levitation system 21 . Silva et al constructed a neural controller to magnetic levitation system and obtained a better control performance than the classical controllers 22 . Wei et al proposed a feedback compensation controller based on backpropagation neural network (BPNN), which lowered the steady-state error in position control of the magnetic levitation ball system 23 .…”

Thiết kế thử nghiệm hệ nâng vật trong từ trường giá rẻ dùng trong nghiên cứu và giảng dạy

“…Ngoài ra, theo nghiên cứu của Silva and Barbosa (2021) Zribi, 2004) có điện áp u là ngõ vào của đối tượng điều khiển (nam châm điện), điện áp này thay đổi sẽ điều khiển được viên bi sắt nâng hạ một khoảng h so với nam châm điện. Khoảng cách h này là ngõ ra của đối tượng điều khiển.…”

Performance optimization of electromagnetic levitation system based on gradient estimator

“…The research and application fields of maglev technology include medium-and-low-speed maglev transportation system (Hu et al, 2021; Min et al, 2011), high-speed maglev transportation system (Wu et al, 2011), magnetic suspension balance system for wind tunnel (Okuizumi et al, 2021; Tashiro et al, 2022), and various high-speed rotating machinery supported by magnetic bearings (Yan et al, 2019; Liu et al, 2016). With the commercial application of various electromagnetic levitation systems, the electromagnetic levitation control technology (Silva and Barbosa, 2021; Yumuk et al, 2021) has been mature.…”