| DC 欄位 |
值 |
語言 |
| DC.contributor | 土木工程學系 | zh_TW |
| DC.creator | 簡宛瑩 | zh_TW |
| DC.creator | Wan-Ying Chien | en_US |
| dc.date.accessioned | 2022-1-24T07:39:07Z | |
| dc.date.available | 2022-1-24T07:39:07Z | |
| dc.date.issued | 2022 | |
| dc.identifier.uri | http://ir.lib.ncu.edu.tw:88/thesis/view_etd.asp?URN=109322043 | |
| dc.contributor.department | 土木工程學系 | zh_TW |
| DC.description | 國立中央大學 | zh_TW |
| DC.description | National Central University | en_US |
| dc.description.abstract | 隨機場是一種能考慮空間變異性的分析方法，基於不同理論其隨機生成的結果也不盡相同。然而鮮少文章探討不同隨機場對地層模型建立的差異及對工程分析的影響。本文針對一潛在液化場址，根據其圓錐貫入試驗結果，分別利用連續隨機場（條件隨機場，簡稱CRF）及非連續隨機場（馬可夫隨機場，簡稱MRF）建立一系列可能的地層模型，並利用信息熵（information entropy, H）量化地層的不確定性。繼之，本文將CRF及MRF生成一系列的地質剖面，在同時考量地層土壤參數（反覆阻抗比）的不確定性後，分別進行液化潛能指數（LPI）的計算，並統計地層中LPI的平均值及變異係數。最後，量化地層不確定性與LPI不確定性之關係。研究結果顯示：（1）地層模型的生成直接受隨機方法影響，MRF模擬之地層分布較CRF模擬之地層分布連續；（2）受第（1）的影響，CRF模擬得地層不確定性較均勻，而MRF的則較不均勻；（3）利用CRF得到的信息熵及LPI不確定性較不具有相關性，而利用MRF得到的信息熵及LPI不確定性則具有正相關性。 | zh_TW |
| dc.description.abstract | A random field is an approach that can represent the spatial variability of soil property. The geological models generated by different random field approaches may yield different results. However, this topic has been seldom discussed. This paper selected two common methods, the covariance matrix decomposition, and the Markov chain Monte-Carlo, as the continuous random field and the discontinuous random field approaches, respectively. The former is referred to as the conditional random field (CRF) and the latter is referred to as the Markov random field (MRF) herein. This paper collected the CPT data on a liquefaction potential site, then calculated the soil behavior type index (Ic) for each borehole. A series of potential geological models then could be generated by each random field approach. This paper introduced the information entropy to quantify the geological model uncertainty. The mean and coefficient of variation of the LPI map could be obtained by analyzing the liquefaction potential with the geological model uncertainty consideration. Finally, the correlation between the geological model uncertainty and the LPI uncertainty could be quantified. The results show that: (1) The geological model generation is mainly affected by random approaches, the stratigraphic configuration simulated by the MRF is more continuous than that simulated by the CRF; (2) The trends of the information entropy map obtained by the MRF and the CRF are similar; however, the spatial variation of information entropy obtained by the CRF is more uniform than that by the MRF; (3) There is an obvious correlation between the information entropy and the LPI uncertainty obtained by the MRF, however, which doesn′t appear in the CRF simulations. | en_US |
| DC.subject | 液化潛能指數 | zh_TW |
| DC.subject | 條件隨機場 | zh_TW |
| DC.subject | 馬可夫隨機場 | zh_TW |
| DC.subject | 地層模型不確定性 | zh_TW |
| DC.subject | 信息熵 | zh_TW |
| DC.subject | liquefaction potential index | en_US |
| DC.subject | conditional random field | en_US |
| DC.subject | Markov random field | en_US |
| DC.subject | geological model uncertainty | en_US |
| DC.subject | information entropy | en_US |
| DC.title | 利用連續及非連續隨機場分析地層與參數變異性對液化潛能指數不確定性的影響 | zh_TW |
| dc.language.iso | zh-TW | zh-TW |
| DC.title | Effect of stratigraphic model uncertainty at a given site on its liquefaction potential index: comparing two random field approaches | en_US |
| DC.type | 博碩士論文 | zh_TW |
| DC.type | thesis | en_US |
| DC.publisher | National Central University | en_US |