ダウンロード数 : ?
タイトルヨミ
モデル テイスウ ノ ユウコウ ケタ オ セッテイシタ リュウシュツ モデル ノ ドウテイ シュホウ ト ソノ テキヨウ レイ
日本語以外のタイトル
An Identification Method of Runoff Model settig a Significant Digit on the Model Parameter and Its Application Cases
ファイル
言語
日本語
著者
福島 晟
森 也寸志
内容記述(抄録等)
This report proposes a modification method of the mathematical optimization technique by the Standard Powell method which is proposed by A. Nagai and M. Kadoya, that is nessary to identfy the runoff model involving parameters which have quantitative relationships of a different order. The outline is as follows.
(1) The modified Standard Powell method is introduced, in which each model parameter is identified to a seached value as sum of an intial value and a value multipled for a constant small term of model parameter. It is considered that a siginificant digit for model parameter is setted to the small term. So, a set of model parameter is searched by a step wise till converging to optimum values.
(2) Three type models are applied to the identification of runoff model. The first type one is the long and short terms runoff model (LST-II), which is developed by Kadoya and Nagai (1988) for analyzing both flood and long term runoff successively. The second type one is the modified LST-H model, in which the effects of distribution of permiable area and impermiable one in the basin is expressed by apllying the structure of the upper zone of first tank LST-II. The third type one is a lumped type model combined the LST-II model with kinematic wave one, named as the lumped KiWS model.
(3) These three type runoff models are adapted to the small exprimental basin at land reclamation area in Masuda City and Hii river basin as large scale one. For the repesentative floods, the model are identified by using the modified SP method. It is shown that the proposed method is improved with respect to evaluate the set of model parameter in the any significant digit designated for each parameter. And it is recognized that the reproducibility of the peak discharge and the observed hydrograph is better for the modified LST-II model than the LST-II model and the lumped KiWS model.
掲載誌名
島根大学生物資源科学部研究報告
3
開始ページ
19
終了ページ
26
ISSN
13433644
発行日
1998-12-20
NCID
AA1112906X
出版者
島根大学生物資源科学部
出版者別表記
Shimane University, Faculty of Life and Environmental Science
資料タイプ
紀要論文
ファイル形式
PDF
著者版/出版社版
出版社版
業績ID
e18398
部局
生物資源科学部
備考
1-12+ / 1996-2007
他の一覧
このエントリーをはてなブックマークに追加