Abstract
Abstract
Active tilt control is a well-established technology in modern railway vehicles, for which currently used control approaches have evolved in an intuitive matter. This paper presents work on a set of novel strategies for achieving local tilt control, i.e. applied independently for each vehicle rather than the whole train precedence approach that is commonly used. A linearized dynamic model is developed for a modern tilting railway vehicle with a tilt mechanism (tilting bolster) providing tilt below the secondary suspension. It addresses the fundamental problems associated with straightforward feedback control, and briefly discusses the current industry norm, which employs command-driven with precedence strategy. Two new advanced schemes are proposed, a model-based estimation approach, and an optimal LQG-based approach, and compared to the command-driven with precedence. The performance of the control schemes is assessed through simulation using a new proposed assessment method.
Keywords
Get full access to this article
View all access options for this article.
