论文部分内容阅读
The spectral representation method(SRM) is widely used to simulate spatially varying ground motions.This study focuses on the approximation approach to the SRM based on root decomposition,which can improve the efficiency of the simulation.The accuracy of the approximation approach may be affected by three factors: matrix for decomposition,distribution of frequency interpolation nodes and elements for interpolation.The influence of these factors on the accuracy of this approach is examined and the following conclusions are drawn.The SRM based on the root decomposition of the lagged coherency matrix exhibits greater accuracy than the SRM based on the root decomposition of the cross spectral matrix.The equal energy distribution of frequency interpolation nodes proposed in this study is more effective than the counter pith with an equal spacing.Elements for interpolation do not have much of an effect on the accuracy,so interpolation of the elements of the decomposed matrix is recommended because it is less complicated from a computational efficiency perspective.
The spectral representation method (SRM) is widely used to simulate spatially varying ground motions. This research focuses on the approximation approach to the approximation approach to the SRM based on root decomposition, which can improve the efficiency of the simulation. The accuracy of the approximation approach may be affected by three factors: matrix for decomposition, distribution of frequency interpolation nodes and elements for interpolation. The influence of these factors on the accuracy of this approach is examined and the following conclusions are drawn. The SRM based on the root decomposition of the lagged coherency matrix exhibits greater accuracy than the SRM based on the root decomposition of the cross spectral matrix. The equal energy distribution of frequency interpolation nodes proposed in this study is more effective than the counter pith with an equal spacing. Elements for interpolation do not have much of an effect on the accuracy, so interpolation of the elements of the decomposed matrix is recommended b ecause it is less complicated from a computational efficiency perspective.