庐山地滑参数反算分析InverseanalysisoftheLushanlandslide.PDFVIP

廬山地滑參數反算分析 賴丞昶 (1) 錢滄海(2) 張光宗 (3) 摘 要 數值分析的參數給定與計算結果準確性有密切的關係，以往藉由經驗或實驗的方式決定地 質材料參數不夠客觀，如有較為客觀的方法找出具有代表性的參數，那將使數值分析的結果更 具價值。本研究利用觀測值與UCODE 程式及類神經網路來校正參數，使模型達到最佳化，模 擬結果與觀測值更為接近。在廬山地滑分析中，摩擦角是對分析結果影響最大的參數，經過 UCODE 程式及類神經網路修正後之參數，皆可改善數值分析的結果。 關鍵字：( 廬山、類神經網路、UCODE 、反算分析) Inverse analysis of the Lushan landslide Cheng-Chang Lai(1) Chang-Hai Chien(2) Kuang-Tsung Chang(3) Engineer(1), Donghan consulting firm, Taichung 411, Taiwan. Assistant Professor(2) (3), Department of Soil and Water Conservation, National Chung Hsing University, Taichung 402, Taiwan ABSTRACT Numerical results closely depend on the material parameters, which are, however, not easy to determine. While they are often subjectively decided by experience or from literature, objective methods should be used to find representative parameters to render more accurate numerical results. This study uses observation data, the program UCODE, and artificial neural networks (ANN) to obtain optimum parameters. In the analysis of the Lushan landslide, the friction angle parameter has the greatest influence on the numerical results. The UCODE and ANN methods show their ability to optimize the model and improve the numerical results. (Keywords ：Lushan, artificial neural networks, UCODE, inverse analysis)                                                         (1)東翰實業有限公司工程師 (2)國立中興大學水土保持學系助理教授 (3)國立中興大學水土保持學系副教授 (通訊作者 e-mail ：changkt@nchu.edu.tw) 185   水土保持學報 43(2) : 185 – 202 (2011) Journal of Soil and Water Conservation , 43 (2) : 185 - 202 (2011) 量做為輸入，材料參數作為輸出，利用 BP