As a series of maglev (magnetically levitated) vehicles travel over a flexible guideway with constant speeds, their acceleration amplitudes will be amplified significantly at resonant or higher speeds. This paper intends to develop a neuro-PI (proportional-integral) controller to control the dynamic response of the maglev vehicles around an allowable prescribed acceleration. The maglev vehicle is simplified as a two-degree-of-freedom (two-dof) moving oscillator controlled by an on-board PI controller and the guideway is modeled as a simply supported beam. Considering the motion-dependent nature of electromagnetic forces working in a maglev system, this study presents an iterative approach to compute the dynamic response of a maglev-oscillator/guideway coupling system based on the Newmark method. The proposed neuro-PI controller is trained using back-propagation neural network in such a way that its PI gains are correlated to the generated data set of moving speeds, mid-span acceleration amplitude of the guideway, and maximum vertical accelerations of maglev oscillators. Numerical simulations demonstrate that a trained neuro-PI controller has the ability to control the acceleration amplitude for running maglev vehicles within an allowable region of prescribed acceleration.
Relation:
Journal of sound and vibration 321(1-2), pp.184-200
