三義火炎山土石流現地監測資料之分析與判識

NCUIR > Engineering College > Graduate Institute of Civil Engineering > Electronic Thesis & Dissertation > Item 987654321/60380

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/60380

Title:	三義火炎山土石流現地監測資料之分析與判識
Authors:	許家銘;Hsu,Chia-Ming
Contributors:	土木工程學系
Keywords:	土石流;次聲;地聲;希爾伯特-黃轉換;debris flows;infrasound;geophones;Hilbert-Huang Transform
Date:	2013-07-30
Issue Date:	2013-08-22 11:35:44 (UTC+8)
Publisher:	國立中央大學
Abstract:	由於近年來全球氣候變遷，極端雨量的出現使土石流發生機率增加，因此本研究經由地聲及次聲監測系統搭配現場影像收集土石流發生時之現場資料，了解土石流發生時之次聲及地聲音頻特性，並分析環境風聲之音頻特性與次聲反應之關係。本研究主要研究之土石流事件為2012/8/2蘇拉颱風引起之火炎山土石流事件。土石流之次聲監測為運用顆粒運動時所散發聲音之特性。故本研究藉由現地事件搭配現場影像找出土石流發生時顆粒撞擊時之音頻特性，並區分由空氣傳播之次聲類型(如土石流引起、風聲引起)與音頻特性之關係，以了解顆粒運動或環境風聲引起發聲機制。　　一般現地低頻噪音成因為風聲、雨聲、地震等因素，實驗及現地資料顯示其特徵頻率皆在5 Hz以下。若以HHT求取解析度更高之特徵頻率約為5-15 Hz，且由此次土石流事件之訊號分析得知地聲所收集之反應較為良好，其特徵頻率概括於10-50 Hz間，且藉由影像系統驗證雙聲系統，得知藉由地聲所測得之反應，分析頻譜以及判定訊號反應之延時，可作為土石流預警系統之根據。 The debris-flow monitoring system setup in Houyenshan Station includes the acoustic sensors, geophone, rain gauge and video recording system. To better identify the characteristics of debris-flow generated signals, the acoustic and seismic signals from both the debris-flow events and environmental noises such as wind gusts, rains, thunders, rockfalls, runoff and earthquakes are analyzed. The acoustic noises such as wind gusts, rains, thunders and earthquakes are generally below 5Hz and with shorter duration (less than 10-100 seconds). The rockfalls are with the frequency range of 30-50 Hz and a short duration of a few seconds. The frequency of seismic data of debris flows are in the range between 10-50 Hz and with a duration greater than 100 seconds. Non-stationary process of the debris flow acoustic signals are demonstrated by using the HHT approach. Frequency rises during incoming surge and falls at the debris-flow tail.
Appears in Collections:	[Graduate Institute of Civil Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	680	View/Open

社群 sharing

Loading...