跳到主要內容

臺灣博碩士論文加值系統

(100.28.0.143) 您好!臺灣時間:2024/07/19 17:20
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

我願授權國圖
: 
twitterline
研究生:張國良
研究生(外文):Zhang, Guo Liang
論文名稱:區域化變數理論在宜蘭地區山坡地土壤力學性質上的應用
論文名稱(外文):An application of the theory of regionalized variables to the mechanical properties of the soils on the slopelands in I-Lan
指導教授:范正成范正成引用關係
指導教授(外文):Fan, Zheng Cheng
學位類別:碩士
校院名稱:國立臺灣大學
系所名稱:農業工程研究所
學門:工程學門
學類:其他工程學類
論文種類:學術論文
論文出版年:1995
畢業學年度:83
語文別:中文
論文頁數:104
中文關鍵詞:區域化變數克氏法複克氏法
外文關鍵詞:Regionalized variablesKrigingCo-Kriging
相關次數:
  • 被引用被引用:2
  • 點閱點閱:97
  • 評分評分:
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
目前台灣地區平地利用幾乎飽和。對於台灣地區山坡地土壤力學性質之研究應將成為往後的課題。
本研究係應用克氏及複克氏系統來評估宜蘭地區山坡地土壤力學性質。因此,幾種傳統的統計方法被用來瞭解區域化變數理論在此個案之應用。
最廣泛使用的三種土壤分類方法分別為USDA, AASHTO和USCS。將本研究地區劃分成網格,而此網格上各點之土壤力學性質就可用上述方法推估而得。利用這些結果在每一網格上之土壤就可被分類出來。根據克氏及複克氏系統分析後,三種土壤分類圖也可繪製出來。由結果得知,運用此兩種系統分析出來的圖有些不同。因此,這兩種系統在統計特性上的差異及其誤差,在本研究中作較深入之探討。在統計分析之後,得到宜蘭地區山坡地土壤力學性質在空間相關性之特性。
本研究亦提出一種決定採樣點數之方法。這種方法是根據採樣點數和推估誤差間之關係,利用非線性迴歸之方法來求得一經驗公式。
利用此公式,就可以用採樣點數來評估推估誤差,反之亦然。這經驗公式導出的過程是很一般化的方法,因此可以應用在其他個案。
本研究的結果期望能對採樣計劃有所幫助,此外,當所需之土壤力學性質不足時,能提供設計的參考。
Taiwan has a very limited land and its plain area has been
nearly fully utilized. So it is necessary to investigate various
mechanical properties of the soil on the slopelands.
This study is to evaluate the mechanical properties of the
soils on the slopelands of the I-Lan area by using the kriging
and co-kriging methods. Therefore, several conventional statistic methods were used to find out the applicability of the theory of the regionalized variables on this project.
There are three most widely used soil classification systems,namely, USDA, AASHTO and USCS. The area of this study is divied into grids. Mechanical properties of the soils on mesh points were estimated. Using this results, the soil on each mesh point was classified. The soil classification maps of the three systems were plotted by using the kriging and co-kriging method.
It is found that the maps by using the two methods were
different. Therefore, differences of the statistical characteristics between the two methods as well as their errors,
were studied. After statistical analyses, spatial relationships
of the mechanical properties of the soils on the slopelands in
I-Lan were found.
In this study, a method for determining sample size is
proposed. This method is based upon the sample size and its
effects on the theoretical errors. An empirical formula is then
obtained by using multiple nonlinear regression techniques.
Using this formula, estimation errors may be evaluated from the sample numbers needed, and vice versa. The formula is empirical,yet it is general. Therefore, it may be applicable to other individual projects.
Results of this study are expected helpful for sampling plans on the slopelands and for designs when the necessary soil
mechanical properties are insufficient.
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top