Comparative analysis of robust inter-controller performance applied to a magnetic levitator

Resumen

This paper presents a comparative analysis of the performance of two robust control strategies applied to a magnetic levitation (Maglev) system, characterized by its nonlinear dynamics and inherent instability. A local controller based on linear ℋ∞ control techniques was designed, evaluated in a closed-loop manner, and contrasted with a previously reported global sliding mode controller under equivalent experimental conditions. The problem formulation included nonlinear modeling of the system, incorporating parametric uncertainties and external disturbances, followed by a linearization process around the equilibrium point for the synthesis of the ℋ∞ controller. Both controllers were implemented in the Magnetic Levitation Systems 33-210 training module from Feedback Ltd., and their performance was evaluated using statistical metrics such as the mean square error (MSE) and the standard deviation (SD) with respect to the operating point. The experimental results indicate that the ℋ∞ controller offers improved stability, lower sensitivity to disturbances, and reduced power consumption, while the sliding-mode controller exhibited oscillations and peak overshoots, attributable to chattering. In conclusion, the ℋ∞ controller is more efficient for this specific system, while sliding-mode control is more suitable in environments with high uncertainty.