Noise propagation analysis with web-based scene modelling

• Main Author: Zhang, Wanyi
• Other Authors: Lee Bu Sung, Francis
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2023
• Subjects: Engineering::Computer science and engineering::Computing methodologies::Computer graphics
• Online Access: https://hdl.handle.net/10356/165357
• Institution: Nanyang Technological University

Singapore is a densely-populated city with a highly interconnected transport network and a heavy reliance on vehicular transport through these roads. As such, traffic noise is one of the leading contributors to noise pollution in urban areas and cities [8], adding on to other sources of sound like industrial noise and various miscellaneous environmental sounds. Residential areas with higher road density also tend to have higher average outdoor sound levels, which is a problem due to the numerous negative health impacts that noise pollution can have on a person. [7] This project aims to improve on an existing noise modelling pipeline. It focuses on more accurately representing the 3D modelled trees used in the pipeline, which are environmental elements that act as noise barriers. It also aims to explore how the species of vegetation can affect the noise propagation in a certain area. It aims to improve the existing pipeline through various means that will be further illustrated below, in order to provide deeper insights on the analysis of the elimination of noise pollution via urban greenery. This paper is split into 3 main parts. 1. Data analysis of different tree species. 2. Modification of existing model to produce more accurate 3D outputs. 3. Noise propagation experiments and evaluations. The first component describes the retrieval of data on different tree species. There is limited information on crown sizes of trees in Singapore, which may be due to constant changes from pruning and other environmental factors, or the difficulty in obtaining such information. The required data hence had to be taken from other research previously done, and cross-referenced with research on the different tree species. The data was then analysed for the second part of the project. The second part utilises the evaluations from the previous part to modify the existing pipeline and produce more accurate representations of various tree species that we have chosen to analyse. The data is exported in a format suitable for use as input in QGIS to visualise the environmental elements in each area. It is then converted into an output that is compatible with the noise propagation prediction tool used in the last part. The last part describes the experiments carried out to examine how different sizes and shapes of noise barriers (i.e. tree species) affects the noise propagation to a receiver building. The experiments are carried out with generalised estimates formulated using several known properties of real-life tree data. This is used to evaluate the suitability of the various tree species in acting as noise barriers.