Wavelet based identification of dominant scales for self-affine road roughness in context of riding comfort in vehicles

Abstract

We present a simulation based study of multiscale roughness of road surfaces and its effect on riding comfort in vehicles, given the fact that characterization of the measured roughness is important to ensure smooth ride. Self-affine fractals are used to simulate typically measured roughness data. Multiscale characteristics of such surfaces are obtained through varied level of spatial resolutions. The hierarchical nature of the multiscale fractal roughness and their role in affecting the riding comfort is investigated herein. Wavelet transform technique is exploited for multiscale decomposition. Roughness is synthesized from cumulation of scales in a multiscale fashion. Single degree of freedom car model is used for characterizing riding comfort parameters in terms of dynamic response of vehicle suspension system subjected to jerks exerted by rough profiles. The dominant scales of roughness governing the comfort parameters are identified through parametric study. It is shown that not all the scales are equally important in deciding the comfort parameters; rather, these parameters remain nearly intact with the inclusion of only a few initial scales. This facilitates multiscale visualization of roughness data and allows representing the profiles in a precise form by excluding the spurious higher scales. Apparently, it gives an economic estimate of the required resolution in the physical measurement for specific purpose.