Publications

Detailed Information

A Study on Noise Sources Identification and Characteristics Deduction of High-speed Vehicles using Sound Visualization and Resolution Optimization : 음장가시화와 분해능 최적화를 통한 고속주행체의 소음원 및 음향특성 도출에 관한 연구

Cited 0 time in Web of Science Cited 0 time in Scopus
Authors

노희민

Advisor
홍석윤
Major
공과대학 산업·조선공학부
Issue Date
2014-02
Publisher
서울대학교 대학원
Keywords
High-speed vehiclesMultiple microphone arrayBeamformingInverse problem optimizationSound characteristics deductionNoise mitigation measures
Description
학위논문 (박사)-- 서울대학교 대학원 : 산업·조선공학부, 2014. 2. 홍석윤.
Abstract
This dissertation presents a study of the localization of major noise sources in a moving vehicle and the deduction of their noise characteristics. In order to localize noise sources in a moving vehicle, a new microphone array was designed on the basis of the beamforming method. The resolution of the array was analyzed from sound-source localization simulations. Validation tests were conducted to identify the locations of noise sources were with a moving vehicle passing-by with horns at specific positions. After the test with the moving vehicle was successfully performed, the process was applied to high-speed trains
the noise map results showed that the majority of noise produced by high-speed trains originates from the front nose, pantograph, inter-coach spacing, and bogie. Then, an improved algorithm was developed for microphone array signal processing to localize the low-frequency noise sources of high-speed trains. The beamforming method does not have high resolution at low frequencies
therefore, it was difficult to derive the aerodynamic noise of the pantograph in the low frequency region. Thus, a deconvolution method was applied to improve the resolution in the low-frequency region, and non-negative constraints and iteration methods were developed to solve the inverse optimization. From the resolution improvement method, the detailed noise sources of the pantograph were derived. Finally, the sound characteristics of the major noise sources in the frequency domain were deduced by transforming the noise maps into the beamforming power spectrum. In order to use the measurement data from the high resolution region, weighting factors were derived with high values in the high resolution region and low values in the low resolution region. The noise characteristics of the front nose were dominant in the low frequency range less than 1000 Hz, and the major difference between the pantograph and its cover occurred at high frequencies greater than 2000 Hz. The noise of the inter-coach spacing section was a tonal sound at frequencies less than 500 Hz. The noise characteristics of the bogies in the leading power car and between car sections were derived
the dominant noise of the bogies was less than 1000 Hz. In conclusion, effective measures that reduce the noise of the main noise sources are proposed on the basis of the results obtained from this research.
Language
English
URI
https://hdl.handle.net/10371/118261
Files in This Item:
Appears in Collections:

Altmetrics

Item View & Download Count

  • mendeley

Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.

Share