查詢結果分析
相關文獻
- 應用遺傳演算法推估橋墩沖刷深度之研究
- 線性軸幅路網接駁系統最適整合區位、路線與排班模式之研究
- 隔板對泥砂導流及局部沖刷之試驗研究
- 克利金空間推估與複迴歸之關係及半變異元函數特性之探討
- Genetic Algorithm Approach for Designing Fir Hilbert Transformers and Differentiators
- 以類神經網路與遺傳演算法解決系統可用度分派問題
- 自我調適的動態排程系統--限制排程、模糊理論和遺傳演算法的應用
- 用無母數的網路學習於臺股認購權證的定價
- 應用遺傳演算法優選大埔水庫排砂操作規線
- 手寫數字辨識模式之建立--結合遺傳演算法與類神經網路
頁籤選單縮合
題 名 | 應用遺傳演算法推估橋墩沖刷深度之研究=Estimation of Pier Scour Depths by Using Genetic Algorithm |
---|---|
作 者 | 王傳益; 陳浩寧; 施漢鵬; 鄭人豪; 張仁瑋; | 書刊名 | 水土保持學報 |
卷 期 | 42:4 2010.12[民99.12] |
頁 次 | 頁377-391 |
分類號 | 441.816 |
關鍵詞 | 遺傳演算法; 局部沖刷; 複迴歸; Genetic algorithms; Local scour; Multiple regression analysis; |
語 文 | 中文(Chinese) |
中文摘要 | 台灣河川坡陡流急,豐枯季節流量差異大,因此河床沖淤變化劇烈,加上橋墩基礎之阻水 作用下,產生局部沖刷,威脅到橋樑的安全。因此瞭解橋墩局部沖刷機制,提出適合之沖刷深 度推估公式,成為當前重要課題。 研究結果顯示,於各評鑑標準下,遺傳演算法(GA, Genetic Algorithms)產出結果之精度絕 大部分優於傳統迴歸方法(Kothyari)。而各項評鑑標準所得之最佳公式與傳統迴歸方式比較其線 性關係及判定係數後,結果顯示以評鑑標準為MAPE 時,GA 所搜尋之優化解於全局搜尋及驗 證與預測上,均較其他評鑑標準準確,其中又以MAPE3 推估所得之結果精度最佳,未來可利 用該式推估均勻橋墩之沖刷深度。 GA 可在一種相對較簡單且穩定的模式中建立,使用遺傳演算法可以有效推估橋墩的沖刷 深。發展GA 只需克服部分困難即可解決許多複雜的問題,且GA 之執行時間較傳統複迴歸分 析方法更為省時。此外,本研究新增各種不同文獻資料做驗證,證明GA 演算結果具適應性及 可靠性。 |
英文摘要 | River bed scour and deposit variation extremely severe because of most rivers are steep and rapid flows, and river discharge is extremely unstable and highly unsteady during different seasons in Taiwan. In addition to the obstruction of piers foundation, it causes local scour and threatens the safety of bridges. Accordingly, to understand the piers scour mechanism and propose an appropriate protection of piers foundation becomes an important topic for study currently. In this study, the Genetic Algorithms (GA) was set up on predicting the local pier scour depth in order to overcome the problem of exclusive and the nonlinear relationships. The observations obtained from nine laboratory in India, Iowa, Japan, French, New Zealand and Florida …etc., were verified by the present model. From the comparison with multiple regression analysis methods, it was found that the scour depth around bridge piers can be predicted efficiently by using the GA. The accuracy of MAPE is higher than other models, where MAPE3 is the best model among the MAPEs, and it will estimate of piers scour depths by using GA in the future. The main advantage of the GA approach over the multiple regression analysis is the relative simplicity. GA model can be set up in a relatively simple and steady way, and applied to more complex problems, and save more time than the traditional multiple regression model. Besides, based on different data set in the literature, it is found that using GA to predict pier scour depth has high applicability and dependability. |
本系統中英文摘要資訊取自各篇刊載內容。