本研究使用波動方程式模擬側向地震下之單樁行為，分析時以間接分析法將完整之土壤與單樁拆解成個別之子結構系統，並依疊加概念分別詮釋不同子結構行為模式，最後加總各系統行為以模擬整體結構反應。研究首先利用集中質塊法結合孔隙水壓模式建立自由場之位移反應，並以此為前置解代入側向樁基波動方程中，求解樁基受液化影響之受震反應。其中單樁周遭土層以p-y曲線模式模擬土壤彈簧，以轉換輻射阻尼係數代入有限差分公式進行分析，樁身剛度並以簡易Bouc-Wen模式模擬其非線性行為。研究另將與有限元素分析軟體Cyclic 1D比較，以證明該項分析之可行性。 研究結果顯示：(1)當改變地震類型、最大地表加速度、孔隙水壓分析模式、地下水位、地盤種類及樁身勁度與阻尼時，皆會影響基樁之破壞情形，其中又以地盤種類與最大地表加速度影響最大；(2)當變更樁長與上部載重時，對於樁體位移、彎矩、剪力皆無明顯之影響；(3)本研究與Hamada（1992）及Ishihara and Cubrinovski（2004）分析結果相似，其樁基礎之最大彎矩易發生於液化層與非液化層交界處；樁身位移之尖峰量發生於樁頂，而樁頂與樁底之相對位移量亦能符合實際現地觀察情形；(4)本研究所建立之孔隙水壓分析模式，可清楚反映樁基礎受震之各個時間下對液化影響產生之變形與受力反應；(5)本研究所建議之簡易分析方法與Cyclic 1D有限元素解所得之趨勢相似，但卻可大量減少分析時間。 The seismic lateral response of the single pile is modeled in this study using the wave equation analysis. The structural system was decomposed into substructures where the corresponding individual responses were solved and integrated to obtain the results. To obtain the solution, the seismic free-field response of the site was computed from the lumped mass analysis and pore water pressure model, and the soil deformations were than imposed onto the pile foundation to conduct the wave equation analysis. The soil springs along the pile are calculated from the p-y curves. They are implemented with the transformed dampers to model the seismic forces. Modified Bouc-Wen model is used to obtain nonlinear pile responses. Validations of the analysis are made with the Cyclic-1D solutions. The observation are summarized as follows: (1) The seismicity source, PGA, depth of ground water table, site stiffness, soil spring and damping, and even the pore water pressure model parameters will all affect the pile responses. Among them, PGA and the site stiffness are the most two significant ones. (2) The changes of pile length and external loads are not sound to the pile displacements and the corresponding pile moments and shears. (3) The solutions from this study were found similar to those of Hamada (1992) and Ishihara and Cubrinovski (2004). The maximum bending moment of the pile occurs at the interfaces between the liquefied layer and the non-liquefiable layers. Peak pile displacement occurs at the pile head. Maximum relative pile displacement between the pile head and pile tip agree well with the field observation. (4) The pore water pressure model can reflect clearly the time-dependence of the trigger of soil liquefaction on seismic response of the pile. (5)The lumped mass analysis of the seismic site responses could provide similar results to those obtained from Cyclic-1D analysis, however it saves more time in computations.
