加權基本不震盪法結合類神經網路與遺傳演算法應用於一維淺水波方程式之求解

Full metadata record

DC Field	Value	Language
dc.contributor.author	李勁頤	en_US
dc.contributor.author	Li, Chin-I	en_US
dc.contributor.author	葉克家	en_US
dc.date.accessioned	2014-12-12T01:39:20Z	-
dc.date.available	2014-12-12T01:39:20Z	-
dc.date.issued	2010	en_US
dc.identifier.uri	http://140.113.39.130/cdrfb3/record/nctu/#GT079716547	en_US
dc.identifier.uri	http://hdl.handle.net/11536/44855	-
dc.description.abstract	本研究以原始高階WENO5算則(Jiang and Shu 1996)為基礎所提出在強震波區域的改進方法：修正平滑指示器WENO5算則(Zhang and Shu 2007)，將之應用於求解一維淺水波方程式，發現在某些情況下，會產生較不好的模擬結果。為改善此模擬結果，本研究以修正平滑指示器WENO5算則的收斂理念為基礎，結合類神經網路(ANN)與遺傳演算法(GA)的原理，透過學習嘗試發展出一套以類神經網路來判斷收斂程度，間接決定WENO算則權重的新算則。最後以一簡單之平滑曲線的微分案例與各種一維明渠流案例進行模擬，比較新算則與各式WENO5算則所得之結果，以評估新算則模擬之收斂性與精確性。	zh_TW
dc.language.iso	zh_TW	en_US
dc.subject	加權基本不震盪法	zh_TW
dc.subject	類神經網路	zh_TW
dc.subject	遺傳演算法	zh_TW
dc.subject	一維淺水波方程式	zh_TW
dc.subject	WENO	en_US
dc.subject	ANN	en_US
dc.subject	GA	en_US
dc.subject	1D shallow-water equations	en_US
dc.title	加權基本不震盪法結合類神經網路與遺傳演算法應用於一維淺水波方程式之求解	zh_TW
dc.title	Applying Weighted Essentially Non-oscillatory Schemes combined with Artificial Neural Network and Genetic Algorithm to Solving 1-D Shallow Water Equations	en_US
dc.type	Thesis	en_US
dc.contributor.department	土木工程學系	zh_TW
Appears in Collections:	Thesis

Files in This Item:

654701.pdf

If it is a zip file, please download the file and unzip it, then open index.html in a browser to view the full text content.