| 姓名 | 吳雯惠(Wen-Hui Wu) | 電子郵件信箱 | s9311613@mail.cyut.edu.tw | |
| 畢業系所 | 營建工程系碩士班(Department and Graduate Institute of Constrction Engineering) | |||
| 畢業學位 | 碩士(Master) | 畢業時期 | 94學年第2學期 | |
| 論文名稱(中) | 土石流發生潛勢與流出土方量推估之研究 | |||
| 論文名稱(英) | The Study on Estimation of Debris Flow Potential and Deposited Volume | |||
| 檔案 |
請遵守中華民國著作權法之相關規定，切勿任意重製、散佈、改作、轉貼、播送，以免觸法。 |
論文使用權限 | 校內外完全公開 | |
| 論文語文/頁數 | 中文/239 | |||
| 摘要(中) | 自921集集大地震之後，中部山區地勢陡峭、地質破碎，加上坡地超限利用之結果，每逢梅雨與颱風季節來臨時，挾帶豐沛雨量，容易發生土石流災害，嚴重威脅坡地居民生命及財產安全。本研究以陳有蘭溪流域內之34條土石流潛勢溪流為研究對象，探討土石流發生潛勢分析與流出土方量推估之研究；主要應用多變量統計分析進行：（1）篩選土石流發生之主要地文因子、材料因子與降雨因子，（2）分析土石流之潛勢，以供溪流整治之參考，（3）進行集群分析，探討各集群土石流影響因子，與（4）建立流出土方量之迴歸模式。 本研究之土石流發生影響因子經篩選後，選定包括集水區面積、形狀係數、平均坡度、水系密度、崩塌面積、地質指標、有效累積雨量與有效降雨強度等八項因子；進一步進行土石流潛勢分析，以費雪區別函數分析與類神經網路判定模式，結果顯示以類神經網路判定之正確率較佳。 經由本研究分析結果，建立集水區面積（A）與崩塌面積（AL）之迴歸方程式AL＝0.0063 A 1.21；另外建立崩塌面積與流出土方量（V）之關係方程式。建議使用流出土方量迴歸方程式 = 349140 AL0.50，作為本研究區域內之流出土方量推估方程式。利用集群分析，依其因子特性進行劃分並探討各群組影響土石流發生因子與流出土方量之關係，且依據流出土方量訂定土石流規模分級。進一步建立三種流出土方量之迴歸模式（1）V =156（A）+525（AL）+ 2189（S）–11576；（2）V ＝262（A）+2489（S）–39240； （3）V ＝154（A）+527（AL）+1716（S）+235（Re）–70924，期望未來評估土石流災害問題時，能快速量度流出土方量，提供相關單位參考依據。 | |||
| 摘要(英) | After the 921
Chi-Chi earthquakes, steep topography, weak geology, and slopeland
overuse caused frequent debris flow in mountain areas in which people
also suffered from debris flow during plum rains and typhoon seasons.
Thirty-four potential debris-flow creeks in the Chen-yu-lan watershed
were studied investigated. Debris flow potential and deposited volume
were estimated, using the following multi-variable statistical
analysis.（1）The effects of topographic, material and rain-related factor
on debris flow were studied investigated.（2）Potential analysis was
perfomed to yield useful information on the debris flow.（3）Factors of
debris flow were studied investigated using cluster analysis.（4）A
regression analysis of the deposited volume was performed. In this work study, eight factors that govern debris flow were studied investigated. These were watershed area, form factor, mean slope, river density, landslide area, geological index, effective accumulated rainfall and effective rainfall intensity. Fisher discriminant analysis and artificial neural networks were used to determine: check debris flow. The resulting corrected rate is more accurate than that obtained using any artificial neural network. The analysis results, the formulas of the landslide area (AL) in terms of the watershed area (A) and the landslide area (AL) in terms of the deposited volume (V ) are as follow AL＝0.0063 A 1.21, = 349140 AL0.50. The variables are grouped by cluster analysis, and the effect of each group on debris flow deposited volume relations. This classification is used to determine debris flow size. Deposited volumes are obtained by regression analysis（1）V =156（A）+525（AL）+ 2189（S）– 11576, （2）V =262（A）+ 2489（S）– 39240,（3）V ＝154（A）+527（AL）+1716（S）+235（Re）– 70924. | |||
| 關鍵字(中) |
|
關鍵字(英) |
| |
| 指導教授 |
| |||