3-D characterisation of road surface textures in TRIAS

摘要

The generation of tyre-road noise is the dominant noise source of road traffic at higher driving speeds. In order to develop noise reduction methods, understanding of the mechanismsinvolved is required. The TRIAS model (Tyre-Road Interaction Acoustic Simulation) predicts tyreroad noise emission via simulation of the interaction between tyre and toad. For the modeling of tyreroad noise, information is needed at least for the whole contact area of a rolling tyre (3-D data). Dataon the road surface texture are normally only available along a line in the driving direction (2-D data, I.e. height and length). Moreover, methods to characterize and measure road texture and tyre tread patterns are not generally accepted. The present paper describes a method to simulate the 3-D texture of road surfaces with a measured 2-D texture profile as a starting point. The method is based on modeling a road surface texture as an auto-regressive filtered white noise sequence. The road texture can be characterized by estimating the parameters of the filter. The modeling is applied to samples of three types of road surface for which the 3-D texture has been measured accurately. The presented method is shown to yield a sufficiently accurate estimate of the 3-D texture by measuring the 2-D profile; a comparison between measured and synthesized 3-D textures indicates a reasonable agreement with respect to the resulting sound levels as predicted with TRIAS. Thus this method may replace the time consuming direct measurement of 3-D textures.