查詢結果分析
相關文獻
- 類神經模糊邏輯法應用於洪水位預報
- 多品質特性系統應用智慧型實驗設計於小型水族箱養殖環境最佳化之研究
- Optimizing the EDM Process with Multi-Response Characteristics by Using the Grey-Based and Fuzzy-Based Taguchi Method
- 應用類神經網路於股票技術指標聚類與預測分析之研究
- 參數即時修正河川洪水預報模式
- 管理元件化資源應用系統--以遠距教學系統為實例
- Vibration Isolation by Using Magnetostrictive Actuator
- 機車防鎖定煞車系統之實作
- 放電加工製程參數最適化之研究
- 模糊集合論在教育評分等級系統之應用
頁籤選單縮合
題 名 | 類神經模糊邏輯法應用於洪水位預報=Application of the Neuro-Fuzzy Logic Algorithm to Flood Stage Forecasting |
---|---|
作 者 | 陳昶憲; 陳建宏; | 書刊名 | 中國土木水利工程學刊 |
卷 期 | 11:2 1999.06[民88.06] |
頁 次 | 頁317-326 |
分類號 | 443.42 |
關鍵詞 | 灰關聯分析; 自組織映射圖網路; 模糊邏輯; 洪水預報; Grey relational analysis; Self-organizing map network; Fuzzy logic; Flood forecasting; |
語 文 | 中文(Chinese) |
中文摘要 | 為了達到即時預報的功能,本文選擇雨量與水位作為洪水預報模式構建參數。由 於集水區雨量站網,屢見儀器故障或出現降雨空間不均勻分佈情況,故若以一般常用流域平 均雨量分析方法推求代表性雨量,則往往出現誤差或無法推求之窘境,故本文先嘗試引用灰 關聯分析,將集水區各雨量站與流量站間關聯性進行排序,藉用各單站雨量與出口水位站水 位建立數組轉換模式;實際應用時可由驗證結果最佳之轉換模式先進行預報,若資料有缺損 則由次高者取代,以此類推。至於轉換模式,本文以模糊邏輯法建立,而模糊邏輯之規則庫 與隸屬函數,則利用類神經網路之自組織映射圖網路推定。文中以烏溪流域之乾峰橋水位測 站之預報為演算實例,進行本文建立之模糊神經洪水位預報法之可行性評估。 |
英文摘要 | In this study, a neuro-fuzzy logic algorithm is develooped for forecasting the river flood stage. To achieve the real-time warning, both rainfall intensity and river stage are chosen as the model variables. Since the uneven rainfall distribution or instrumental error of aguging stations may affect significantly the resulting representative rainfall obtained by the socalled weight-average approach, a remedy based on grey relational analysis is proposed to generate a relative sequence of rainfall records in different locations among the same basin. According to this relative sequence, several transfer models constructed by fuzzy logic method, in which the self-organizing map of neural network is used in learning the rule and membership function, are proposed for forecasting the river stages. An assessment of the feasibility of proposed algorithm on flood forecasting is evaluated by using the data of Chin-Fon bridge, an upstream gauging station of Wu-Shis basin. |
本系統中英文摘要資訊取自各篇刊載內容。