论文部分内容阅读
结合自由场地土振动台试验,对比了随机子空间迭代(SSI)方法、经验模式分解结合希尔伯特变换(EMD+HT)方法和频域分解(FDD)方法识别场地土卓越频率、阻尼比的适用性,以及不同激励类型、不同土层深度和大阻尼比情况对识别精度的影响。结果表明:对于随机激励,当阻尼比较小时,上述3种方法均能准确识别出场地土卓越频率;而大震时土体卓越频率、阻尼比与振动过程耦合,此时经验模式分解结合希尔伯特变换方法可以给出合理的识别结果。
Combined with free-field soil shaking table test, the stochastic subspace iteration (SSI) method is compared. Empirical mode decomposition combined with Hilbert transform (EMD + HT) and frequency domain decomposition (FDD) The applicability and the influence of different excitation types, different soil depths and large damping ratios on the recognition accuracy. The results show that for the random excitation, when the damping ratio is small, the above three methods all can accurately identify the excellent earth frequency of the site; while the earth’s excellent frequency and damping ratio of large earthquakes are coupled with the vibration process, the empirical mode decomposition combined with Hill Burt transformation method can give a reasonable recognition result.