頁籤選單縮合
題 名 | 適用於擬動態試驗之具數值消散特性的外顯式積分法=A Family of Dissipative Explicit Methods for Pseudodynamic Testing |
---|---|
作 者 | 張順益; | 書刊名 | 中國土木水利工程學刊 |
卷 期 | 10:3 1998.09[民87.09] |
頁 次 | 頁495-503 |
分類號 | 440.11 |
關鍵詞 | 數值消散; 外顯式積分法; 擬動態試驗; Numerical dissipation; Explicit method; Pseudodynamic test; |
語 文 | 中文(Chinese) |
中文摘要 | 使用具有數值消散特性的逐步積分法已被認為是擬動態試驗上克服因高頻振態 所引起誤差累積傳播效應的有效方式,然而,現今具有數值消散特性的逐步積分法幾乎都是 內隱式積分法,而內隱式積分法在擬動態試驗上的應用較為繁複且不易,因而較少被採用。 本文將詳細推導現今唯一具有數值消散特性且仍保有二階精確度的外顯式積分法。此積分法 非常適用於擬動態試驗上,用以克服試驗時由於數值誤差或試驗誤差所引起不正確的高頻振 態反應。 為了突顯該積分法的數值消散特性,Newmark 外顯式γ - 消散法亦作比較研究。 這些比較將包括數值阻尼比,相對週期誤差以及頻譜半徑隨時間步長與自然週期之乘積而改 變的情形。最後應用此積分法完成兩個數值釋例。 |
英文摘要 | It has been recognized that numerical dissipation is an effective way for suppressing the spurious high-frequency modes. However, most dissipative integration methods are implicit algorithms for which the implementation for application to pseudodynamic tests appears to be more complicated than the explicit ones. A family of dissipative explicit methods is proposed herein, which possesses the same second-order accuracy as the basic property for the commonly used integration methods. These method are suitable for pseudodynamic tests due to existence of numerical dissipation for eliminating the spurious high-frequency response. For comparison, the Newmark explicit method with γ -dissipation is also investigated. Factors investigated include the variations of numerical damping ratio, relative period error and spectral radius with respect to the product of the natural frequency and the size of time step. Two numerical examples are provided to demonstrate the effect of numerical dissipation using the proposed algorithms. |
本系統中英文摘要資訊取自各篇刊載內容。