Traffic-related noise has been increasing steadily. Noise barriers are one of the main tools used for noise abatement, and there is still potential for optimization and improvement of the acoustic performance by employing non-standard designs. Simulations are a cost-efficient tool for predicting and planning new noise barrier solutions. The following paper studies some non-standard barrier shapes with particular focus on the formation of a virtual soft-plane for some frequencies. Destructive diffraction from the top edge of the barrier is used in order to optimize the shielding effect of the barrier. Through the use of 2D-BEM simulations different barrier profiles and their effect of shielding are studied. The focus is to obtain useful shielding in the farfield region with intelligent shapes thus permitting a reduction of the barrier height.
In this article, statistical modelling approaches for tyre/road noise levels (according to the CPX method) based on continuous 3D texture measurements are presented. The main focus is to estimate if viable correlations of interpretable 3D texture parameters with frequency-dependent CPX levels can be found. Therefore, a set of descriptive 3D texture parameters is introduced. Two different modelling approaches, focussing on linearity in order to gain interpretability, are shown. First, a linear model is calculated based on the first principal components of the texture parameters, and second, a random forest regression approach is performed on the direct texture parameters. As a principal component analysis of the CPX measurement data reveals three highly correlated frequency ranges, both approaches are calculated independently for low, mid and high frequency bands. Both models show good performance in the low and mid frequency range, whereas the accuracy in the high frequency range declines. Due to the focus on linearity, both approaches result in comparable statistical benchmark parameters.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.