Acoustic amenity analysis for high-rise building along urban expressway: Modeling traffic noise vertical propagation using neural networks

摘要

Traditional noise propagation models are built for particular height of building. Residents living in high-rise buildings are severely affected by noise. The main objective of this study is to analyze acoustic amenity and develop noise analysis model for high-rise buildings, especially those along urban expressway. The study was carried out in three stages. Firstly, noise survey on different floors of high-rise buildings along free flow expressway was conducted using noise measuring equipment. Furthermore, C-weighted network was used in conjunction with A-weighted network to determine low frequency noise, which is perhaps less noticed for most survey. Secondly, the noise indicators change rule on vertical plane was explored. Combining change rule with frequency spectrum analysis, the acoustic amenity for high-rise buildings was carried out deeply and strictly. In the last stage, traffic noise prediction model for high-rise building along expressway was constructed with neural network. A point worth emphasizing is that the comparison between developed model, FHWA model and measurement shows that the proposed model fits well with measurement as compared to FHWA model at 5% significance level. The developed method could be used as a tool for acoustic amenity analysis and model building. It would be possible to provide reference for urban expressway and building design. (C) 2017 Elsevier Ltd. All rights reserved.