Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/86182
DC FieldValueLanguage
dc.contributorDepartment of Mechanical Engineering-
dc.creatorWong, Hon-yiu-
dc.identifier.urihttps://theses.lib.polyu.edu.hk/handle/200/4353-
dc.language.isoEnglish-
dc.titlePrediction models for the acoustic performance of noise barriers-
dc.typeThesis-
dcterms.abstractNoise barriers are the common acoustic measures which are used to minimize the disturbance of traffic noise to residents. It is essential to have an accurate prediction scheme for barrier designs. Poor prediction schemes will lead to an undesired performance of barriers or wastage of money for over design. The prediction schemes used currently are only suitable to a barrier with a very simple configuration. Some sophisticated numerical schemes are available for barrier problems with complex configurations, but these schemes require high computational efforts. The main objectives in this thesis are to develop prediction schemes for present barrier problems provided that the prediction schemes shall have a good balance between prediction accuracy and computational efficiency. A study of the prediction models for different types of barriers has been investigated. Gaps are found in some new built barriers. The gaps are left in barriers for emergency access and maintenance for instance. The presence of gaps may significantly reduce the overall insertion loss of a barrier. In contrast, small gaps on a barrier are reported as a preferential frequency filters which enhance the insertion loss. However, no simple computational schemes are available to predict the effects of the barrier gaps. There are often many conflicting design factors that have to be considered when specifying a barrier. Acoustic engineer has to balance the barrier performance and the barrier appearance. Thus there have been pressures on the acoustic engineer to provide ever more creative solutions to enhance the overall insertion loss provided by barriers without increasing their heights. Using cantilevered barriers, tilted barriers, cranked barriers, and non-straight top profile barriers are the examples of current strategies to improve the barrier performance. However, the performance of these special design barriers is only approximated by some empirical formulae. Thus, uncertainties in estimating barrier performance are introduced. More accurate prediction schemes are in need. Evaluate barrier performance periodically is an important issue for assuring the efficiency of barrier. The International Standards state methods for evaluating the performance of a built barrier. It is necessary to find an equivalent site for recovering the sound levels before the barrier has been built. However, it is hard to find an exactly equivalent site under the same traffic. So the applicability of stated methods is always in questioned. Improvement on the current in-situ determination method for barrier performance is in needed. The research in this thesis aims to handle the problems stated before. The first addressed research topic is barrier leakage. An approximate boundary integral method is used to model the barrier leakage problems due to horizontal and vertical gaps. Straightforward implemented integral formulae are presented and these formulae are used to predict the leakage effects. A ray method is also presented for the prediction of leakage due to horizontal gaps. This ray method is a more computational efficient method than using integral formulae with compatible accuracy. The second research topic is to predict the sound levels behind tilted barriers and cranked barriers of finite length. An approximate boundary integral method is used. The theories on sound propagation inside wedge are introduced to approximate boundary integral method for accounting the multiple reflections of sound between the barrier surfaces and the ground surface. A hybrid ray-BEM model is developed of which a correction factor for edge effects based on geometrical diffraction models is added to a two-dimensional BEM model, for estimating the sound levels behind a multiple edges barriers of finite length. Furthermore, the concepts of using teeth top barriers are introduced and an approximate boundary integral method is used to predict the diffracted sound by such teeth top barriers. It is shown that teeth top barriers can significantly improve the barrier performance. Lastly, an improved scheme of in-situ determination of barrier performance has been developed. Semi-empirical formulae for estimating the sound levels in "Before" situation have been derived by approximating the Weyl-van der Pol formula. It is shown that better accuracy in in-situ determination of barrier performance is achieved after implementing the proposed schemes. The theories and formulae proposed in this thesis are validated by conducting indoor experiments in an anechoic chamber, performing full scaled outdoor measurements, and comparing to benchmark numerical schemes. The proposed theories are shown to be accurate and useful to encounter the addressed problems.-
dcterms.accessRightsopen access-
dcterms.educationLevelPh.D.-
dcterms.extentxxii, 324 leaves : ill. (some col.) ; 30 cm-
dcterms.issued2004-
dcterms.LCSHHong Kong Polytechnic University -- Dissertations-
dcterms.LCSHNoise barriers -- Evaluation-
dcterms.LCSHSound -- Measurement-
Appears in Collections:Thesis
Show simple item record

Page views

43
Last Week
2
Last month
Citations as of Mar 24, 2024

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.