透過您的圖書館登入
IP:18.191.228.88
  • 學位論文

應用貝氏最大熵法於臺北盆地水文地質推估

Estimation of Lithological Classification in Taipei Basin: A Bayesian Maximum Entropy Method

指導教授 : 余化龍

摘要


在自然環境科學和工程應用等研究領域中,在分析地質、水文、地下水或是洪旱災時,對研究區域的地質組成岩性有一定的了解,能夠使研究方向和觀念更加正確,進而幫助研究順利進行。但由於受到許多自然和人為的條件侷限,導致實際採樣和觀測資料相當有限,許多研究僅能使用有限的觀測資料進行分析。為有效了解未知空間點的岩性分布,需透過空間統計方法來對未知點進行推估,傳統上有許多地理統計方法應用於此,而本研究應用的貝氏最大熵法(Bayesian Maximum Entropy, BME method)為其中之一新興的時空間地理統計方法。其結合數值模式方法與資料導向方法,可以同時考慮時間和空間得相關性,並以貝氏條件機率的概念為基礎,結合物理知識和其他不確定性資料以增強推估資訊,同時對序率資料與空間資料進行推估分析。而岩性分類的地質資料屬於離散分布的類別型資料,因此本研究應用「類別型貝氏最大熵法(Categorical BME)」,並同時考慮限制式可能存在的不確定性,嘗試放鬆限制式的不確定性範圍後進行模式收斂,以針對岩性類別資料進行推估分析。透過地質鑽探所取得之有限的岩心資料,建立完整的三維類別型貝氏最大熵法岩性推估模式,推估台北盆地之三維岩性分布結果,期望可供台北盆地地質未來相關研究之參考。

並列摘要


In environmental scientific applications and studies, we have to understand the geological lithological composition in the study area. Because of some restrictions of the situation in reality, only limited amount of data can be acquired. In order to find out the lithological distribution in the study area, many geological spatial statistical methods used to analyze the lithological composition on unsampled points or grids. This study applied the Bayesian Maximum Entropy (the BME method), which is a new method in the geological spatiotemporal statistics field. The BME method can compute the spatiotemporal correlation of the data, and integrate not only the hard data but the soft data to improve the results of estimation. The data of lithological classification is discrete categorical data. Therefore, this research attempt to apply Categorical BME to establish a complete three-dimensional Lithological estimation model. And try to regularize the maximum entropy density estimation. Apply the limited hard data from the cores and the soft data generated from the geological dating data and the virtual wells to estimate the three-dimensional lithological classification in Taipei Basin.

參考文獻


9. Berger, A. (1997). The Improved Iterative Scaling Algorithm: A Gentle Introduction: School of Computer Science, Carnegie Mellon University.
10. Bogaert, P. (2002). Spatial prediction of categorical variables: the Bayesian maximum entropy approach. Stochastic Environmental Research and Risk Assessment (SERRA), 16(6), 425-448. doi: 10.1007/s00477-002-0114-4
12. Bogaert, P., & Fasbender, D. (2008). Nonlinear Spatial Prediction with Non-Gaussian Data: A Maximum Entropy Viewpoint. geoENV VI – Geostatistics for Environmental Applications, 445–455.
13. Chernick, M. R. (2008). Bootstrap methods : a guide for practitioners and researchers (2nd ed.). Hoboken, N.J.: Wiley-Interscience.
14. Christakos, G. (1990). A Bayesian Maximum-Entropy View to the Spatial Estimation Problem. Mathematical Geology, 22(7), 763-777. doi: Doi 10.1007/Bf00890661

延伸閱讀