EUVIMEDEUVIMED entdecken
Open Access · CC BY-SA 4.0

Wavelet Neural Network-Based Controller Design for Magnetic Levitation System

Abdulla Ibrahim Abdulla, Mohammed Qasim, Mohammed Almaged

Journal of Robotics and Control (JRC) · 17.05.2025 · 6 (3) · 1273-1282

DOI: 10.18196/jrc.v6i3.26237 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Robotics and Control (JRC), online erschienen am 17.05.2025. Verfasst von Abdulla Ibrahim Abdulla und 2 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref). Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The magnetic levitation system (MLS) poses a substantial control challenge owing to its intrinsic instability and pronounced nonlinear dynamics. The implementation of robust control methodologies is imperative to guarantee stable operational performance, particularly in environments characterized by external disturbances and parametric uncertainties. This study investigates the development of a PID-like control strategy for a magnetic levitation system (MLS), employing WNN architecture. The parameters of the proposed controller are optimized by employing Fick's Law Algorithm (FLA). The optimization process utilizes a cost function that comprises a weighted sum of the Integral Time-weighted Square Error (ITSE), Integral Time-weighted Absolute Error (ITAE), maximum overshoot (MO), and minimum undershoot (MU). This multi-objective cost function enables a comprehensive evaluation of the controller's performance across various criteria. A square wave reference signal is employed to conduct the optimization process, presenting a challenging test case for control system performance due to its abrupt transitions. The efficacy of the proposed controller is evaluated through a comparative analysis with a conventional PID controller. Comparative simulations are conducted employing three distinct reference trajectories: step, sinusoidal, and square waves. These diverse trajectories provide a comprehensive evaluation of the controller's performance. To assess the robustness of the proposed controller, simulations are conducted within the MATLAB/Simulink environment, subjecting the MLS model to both external disturbances and parametric uncertainties. The developed controller exhibits superior performance and robustness characteristics in comparison to the conventional PID controller. It effectively attenuates the detrimental impact of both parametric uncertainties and external disturbances, while concurrently maintaining a high degree of performance accuracy in terms of overshoot, steady-state error, and energy consumption.

Zitieren

APA

Abdulla, A. I., Qasim, M., & Almaged, M. (2025). Wavelet Neural Network-Based Controller Design for Magnetic Levitation System. Journal of Robotics and Control (JRC), 6(3), 1273-1282. https://doi.org/10.18196/jrc.v6i3.26237

Vancouver

Abdulla AI, Qasim M, Almaged M. Wavelet Neural Network-Based Controller Design for Magnetic Levitation System. Journal of Robotics and Control (JRC). 2025;6(3):1273-1282. doi:10.18196/jrc.v6i3.26237

BibTeX
@article{Abdulla2025,
  author = {Abdulla, Abdulla Ibrahim and Qasim, Mohammed and Almaged, Mohammed},
  title = {Wavelet Neural Network-Based Controller Design for Magnetic Levitation System},
  journal = {Journal of Robotics and Control (JRC)},
  year = {2025},
  volume = {6},
  number = {3},
  pages = {1273--1282},
  doi = {10.18196/jrc.v6i3.26237}
}
Kostenlos starten →