Neuro-Fuzzy Control of Railcar Vibrations Using Semiactive Dampers

摘要

This article describes a new approach to reducing vertical vibrations in a 70-ton railcar using a neuro-fuzzy controller and a rnagnetorheological (MR) damper. A semiactive control technique is developed for a two-degree-of-freedom quarter car model of the railcar that has an installed MR damper. A fuzzy controller in real time continuously updates damping properties of the device. The controller uses feedback acceleration of the freight mass to specify a voltage signal to the MR damper. Correlations between acceleration (controller input) and voltage applied to the MR damper (controller output) are developed using Neuro Fuzzy Controller (NEFCON). To assess effectiveness of the semiactive control scheme, responses of the railcar to various haul conditions are compared with those for uncontrolled and for passive operating conditions of the MR damper. Results indicate that semiactively controlled MR dampers can reduce vibrations to acceptable levels provided that sufficient force capacity can be supplied by the damper.