H∞ mixed sensitivity optimization for high speed tilting trains

Abstract

The industrial norm of tilting high speed trains, nowadays, is that of Precedence tilt (also known as Preview tilt). Precedence tilt, although succesfull as a concept, tends to be complex (mainly due to the signal interconnections between vehicles and the advanced signal processing required for monitoring). Research studies of early prior to that of precedence tilt schemes, i.e. the so-called Nulling-type schemes, utilized local-per-vehicle signals to provide tilt action (this was essentially a typical disturbance rejection-scheme) but suffered from inherent delays in the control). Nulling tilt may still be seen as an important research aim due to the simple nature and most importantly due to the more straightforward fault detection compared to precedence schemes. The work in this paper presents a substantial extension conventional to robust H∞ mixed sensitivity nulling tilt control in literature. A particular aspect is the use of optimization is used in the design of the robust controller accompanied by rigorous investigation of the conflicting deterministic/stochastic local tilt trade-off.