Rock Mass Permeability Identification by Integrating the Techniques of Geological Features and Logistic Regression

View Statistics Email Alert RSS Feed

Information

Details

Project title

Code/計畫編號

MOST109-2116-M019-004

Translated Name/計畫中文名

結合地質特徵與羅吉斯迴歸分析法於岩體滲透性的辨識

Project Coordinator/計畫主持人

Shih-Meng Hsu

Funding Organization/主管機關

National Science and Technology Council

Department/Unit

Department of Harbor and River Engineering

Website

https://www.grb.gov.tw/search/planDetail?id=13547361

Year

2020

Start date/計畫起

01-08-2020

Expected Completion/計畫迄

31-07-2021

Bugetid/研究經費

848千元

ResearchField/研究領域

地球科學

裂隙岩體滲透性參數的調查，與臺灣山區現行面對許多與地下水相關之工程實務問題有關，此參數通常在場址選擇或施工規劃前，需要針對它進行瞭解及取得，此工作更對於後續的工程設計、施工災害預防及作為候選場址之決定提供重要關鍵的資訊，但過往由於計畫經費限制多以點狀調查為主，如何以更有效率與低成本的方式辨識候選場址的岩體滲透性質是尚待解決問題?本計畫蒐集全台各流域裂隙岩體既有探勘之地質與水文地質試驗資料，以此跨域與多岩性的資料為基礎，嘗試應用羅吉斯迴歸分析方法，結合與岩體滲透性質具高度相關性的地質特徵，建構適當場址發生機率與選定之地質特徵因子間的關係，最後產出以地質特徵預測達成岩體滲透係數門檻值成功機率之迴歸模式，且應用統計的指標來評估所建立模式的正確性，冀望藉此模式來評估候選場址適合度，提供相關開發單位工程選址階段參考。 Investigations of fractured rock permeability are closely related to a number of practical engineering problems associated with groundwater issues in Taiwan mountainous areas. This parameter needs to be understood and acquired before selections or planning of a candidate site. The investigated results can further provide crucial information for engineering design, the precaution of construction disasters, and selections of candidate sites. However, the budget for past projects was limited to plan a detailed investigation of the hydraulic properties of a borehole. Therefore, a question regarding how to identify rock mass permeability of a candidate site with high efficiency and low cost is still required to be answered. This project collects existing geological and hydrogeological data of Taiwan mountainous areas with cross-regional and multi-lithological features. Based on these data, this study integrates the techniques of geological features and logistic regression to explore relationships between selected geological features (independent variables) and the probability of passing the permeability threshold (a binary variable). And then, a logistic regression model will be produced and can be used to predict the probability of output (passing permeability threshold) in terms of input (selected geological features). Finally, the developed model will be evaluated by a few statistical tests. The verified model can assist in evaluating the feasibility of a candidate site, which is a basis for selecting an engineering site of a construction company during the phase of site-selection.

Keyword(s)

裂隙岩體
滲透性
地質特徵
羅吉斯迴歸
fractured rock
permeability
geological features
logistic regression

DSpace CRIS

Rock Mass Permeability Identification by Integrating the Techniques of Geological Features and Logistic Regression

Details

Description