Lyapunov Exponents and Information Dimension of Nonlinear Railway Wheelsets Incorporating Randomness.

Abstract- This article presents a numerical simulation that aims to examine the effect, of randomness in forward speed, in lateral clearance (dead band) and in both, on the dynamic behaviors of single- and two-axle railway wheelsets. Randomness is represented by pseudo-random numbers. They are incorporated into the dynamic models of the wheelsets. Subsequently, the temporal average of Lyapunov exponents is determined. The ensemble average of these exponents is then used to compute the information dimension. It is found that the introduction of small to moderate randomness does not necessarily lead to a chaotic response in the wheelset; in fact, the presence of small to moderate randomness may suppress chaotic response otherwise existing.