当前位置: X-MOL 学术Soil Dyn. Earthq. Eng. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Mapping surface wave dispersion uncertainty in Vs Profiles to VS,30 and site response analysis
Soil Dynamics and Earthquake Engineering ( IF 4 ) Pub Date : 2020-11-01 , DOI: 10.1016/j.soildyn.2020.106298
Narayan Roy , Ravi S. Jakka

Abstract The article studies the propagation of dispersion uncertainty in Vs profiles to VS,30 and linear soil amplification, and at the same time, it also attempts to infer some insights on how the inversion non-uniqueness plays a role in the extracted Vs profiles and subsequent analyses. Three synthetic and two field studies have been presented for this purpose. Target dispersion curve has been assigned with an uncertainty estimate and inversion has been performed using Neighbourhood algorithm. Profiles have been selected from different misfit ranges and combined into one in order to map dispersion misfit with VS-misfit, VS,30 and peak soil amplification. The term VS-misfit has been introduced for synthetic cases to represent the deviation of an extracted VS profile after inversion from the true solution using a single value. The results clearly depict that the range of variation of VS-misfit, VS,30 and peak amplification increases as the dispersion misfit increases. This implies that if the minimum misfit achieved during inversion is higher, the inversion non-uniqueness might play a bigger role in the subsequent analyses. As the dispersion misfit increases, more number of solutions are found to exist at very similar misfit values and the non-uniqueness of inversion manifests more noticeably in the subsequent analyses. Standard deviation is found to increase with the increase in misfit ranges. In terms of COV, VS-misfit exhibits the highest variation, whereas the variations get reduced in VS,30 and in peak amplification.

中文翻译:

将 Vs Profiles 中的表面波色散不确定性映射到 VS,30 和现场响应分析

摘要 本文研究了 Vs 剖面中频散不确定性向 VS,30 和线性土壤放大的传播,同时也试图推断出反演非唯一性如何在提取的 Vs 剖面中发挥作用的一些见解和后续分析。为此目的提出了三项综合研究和两项实地研究。目标色散曲线已分配有不确定性估计,并已使用邻域算法进行反演。从不同的失配范围中选择剖面并将其合并为一个,以便用 VS-失配、VS,30 和峰值土壤放大来绘制分散失配。术语 VS-misfit 已被引入合成案例,以表示使用单个值从真实解反演后提取的 VS 轮廓的偏差。结果清楚地描述了 VS-misfit、VS,30 和峰值放大的变化范围随着色散失配的增加而增加。这意味着如果反演期间实现的最小失配更高,反演非唯一性可能在后续分析中发挥更大的作用。随着色散失配的增加,发现更多的解存在于非常相似的失配值,反演的非唯一性在随后的分析中更加明显。发现标准偏差随着失配范围的增加而增加。就 COV 而言,VS-misfit 表现出最高的变化,而变化在 VS,30 和峰值放大中减少。这意味着如果反演期间实现的最小失配更高,反演非唯一性可能在后续分析中发挥更大的作用。随着色散失配的增加,发现更多的解存在于非常相似的失配值,反演的非唯一性在随后的分析中更加明显。发现标准偏差随着失配范围的增加而增加。就 COV 而言,VS-misfit 表现出最高的变化,而变化在 VS,30 和峰值放大中减少。这意味着如果反演期间实现的最小失配更高,反演非唯一性可能在后续分析中发挥更大的作用。随着色散失配的增加,发现更多的解存在于非常相似的失配值,反演的非唯一性在随后的分析中更加明显。发现标准偏差随着失配范围的增加而增加。就 COV 而言,VS-misfit 表现出最高的变化,而变化在 VS,30 和峰值放大中减少。发现更多的解存在于非常相似的失配值下,反演的非唯一性在随后的分析中更加明显。发现标准偏差随着失配范围的增加而增加。就 COV 而言,VS-misfit 表现出最高的变化,而变化在 VS,30 和峰值放大中减少。发现更多的解存在于非常相似的失配值下,反演的非唯一性在随后的分析中更加明显。发现标准偏差随着失配范围的增加而增加。就 COV 而言,VS-misfit 表现出最高的变化,而变化在 VS,30 和峰值放大中减少。
更新日期:2020-11-01
down
wechat
bug