查詢結果分析
相關文獻
- 烏溪流域洪流時序分析
- 洪流之多變量時序分析
- 臺灣地區家禽霍亂之發生
- Fault-toleration by Duplication and Debugging for Distribution Real-time Systems
- 以Verilog PLI發展之1-Context非線性延遲計算器
- 高性能數位VLSI之時序分析與模擬技術
- 「事件時序分析法」(Event History Analysis)簡介
- Gone with the Wind?: Legitimacy and Governor Patten's Popularity in the Transition Period
- Air Pollutants and Suicide Rate: A Cross-sectional Time Series Analysis
- 精微沖鍛自動化系統開發
頁籤選單縮合
題名 | 烏溪流域洪流時序分析=Time Series Analysis of Flood Discharge in Wu-Shi Basin |
---|---|
作者 | 陳昶憲; 黃偉民; 朱益辰; | 書刊名 | 臺灣水利 |
卷期 | 45:3=179 1997.09[民86.09] |
頁次 | 頁72-82 |
分類號 | 443.42 |
關鍵詞 | 時序分析; 轉換函數噪音模式; 線性轉換函數法; Time series analysis; Transfer function noise model; Linear transfer function method; |
語文 | 中文(Chinese) |
中文摘要 | 本文應用時間序列模式單變量AMIMA及多元變量TFN,分別針對烏溪流域大肚橋及 其上游三測站之流量進行分析及比較,結果顯示加入上游測站之TFN模式,其大肚橋流量預 測結果顯然此單一測站ARIMA模式更其準確性,其差異尤其在多階前進預測上更為明顯,其 於三階預測之相關係數平方R□,多元變量TFN模式仍可達到0.79以上,而單變量ARIMA模 式則僅能達到0.64。故從預報能力而言,本文建立之多元變量時序模式對預測結果明顯的有 較佳的趨勢掌握,且其具實用上之高階預測能力。 |
英文摘要 | In this paper, univariate ARIMA and multivariate TFN time series models are applied to analyze and forecast the catchment flood discharge on the Wu-Shi basin. The gauging stations which were discussed in this article, included the downstream gauging station, Da-Du bridge and the upstream gauging stations, His-Nan bridge, Nan-Gan bridge and Gan-Feng bridge. The forecasting result of Da-Du bridge was shown that TEN model is much precise than ARIMA model, especially on multi-steps ahead forecasting. The R-square value can be attained 0.79 for multivariate TFN model even if the lead step of forecast is third, but it only attain 0.64 for univariate ARIMA model. So consider the usefulness of forecasting, multivariate time series model is obviously more superior on the tendency control, and it possess practical function by multi-steps ahead forecasting. |
本系統之摘要資訊系依該期刊論文摘要之資訊為主。