基于多源数据融合的土体空间参数不确定性缩减  

Spatial uncertainty reduction of soil parameters based on multi-source data fusion

作　　者：贾唯龙 常鹏飞 李亚军 钱铖 郭祺 李瑞杰 傅中志[5] 张彬[1,2] JIA Weilong;CHANG Pengfei;LI Yajun;QIAN Cheng;GUO Qi;LI Ruijie;FU Zhongzhi;ZHANG Bin(School of Engineering and Technology,China University of Geosciences(Beijing),Beijing 100083,China;Engineering and Technology Innovation Center for Risk Prevention and Control of Major Project Geosafety,Ministry of Natural Resources,Beijing 100083,China;Coal Geological Geophysical Exploration Surveying&Mapping Institute of Shanxi Province,Jinzhong 030600,China;Shanxi Road and Bridge Third Engineering Co.,Ltd.,Xinzhou 034000,China;Nanjing Hydraulic Research Institute,Nanjing 210024,China)

机构地区：[1]中国地质大学(北京)工程技术学院,北京100083 [2]自然资源部重大工程地质安全风险防控工程技术创新中心,北京100083 [3]山西省煤炭地质物探测绘院有限公司,山西晋中030600 [4]山西路桥第三工程有限公司,山西忻州034000 [5]南京水利科学研究院,江苏南京210024

出　　处：《水资源与水工程学报》2024年第3期183-191,共9页Journal of Water Resources and Water Engineering

基　　金：资源环境与灾害监测山西省重点实验室开放课题(2023-S01);中央高校基本科研业务费项目(2652019321)。

摘　　要：土体参数存在空间变异性,导致结构响应具有不确定性。忽视这种不确定性或采用偏差参数可能导致工程安全问题,甚至引发工程灾害。现场测量数据可用于改进对基坑开挖等岩土工程中结构响应的估计,通过SPT等直接测量方法得到的勘察数据与土体刚度参数直接相关,在这些数据的基础上,基于Kriging法约束随机场的条件模拟可用于改进对参数空间分布的估计。此外,当监测数据与位移等土体结构性能或响应相关时,利用集合卡尔曼滤波器(EnKF)反分析也可降低土体参数的不确定性。本文采用多源数据融合的形式将直接和间接两种方法结合在一起进行土体开挖分析,结果表明:两种方法结合使用可以显著降低土体参数的不确定性,从而降低结构响应中的不确定性。The spatial variability of soil parameters leads to uncertainties in structural response.Ignoring these uncertainties or using biased parameters may result in engineering safety issues and even cause engineering disasters.Field measurement data can improve the estimation of structural responses in geotechnical engineering,such as foundation pit excavation.Investigation data obtained from direct measurement methods like the standard penetration test(SPT)are directly related to soil structural parameters.Based on these data,conditional simulation constrained by the Kriging method can improve the estimation of the spatial distribution of parameters.Additionally,when monitoring data are related to soil structural performance or responses,such as displacement,inverse analysis using the ensemble Kalman filter(EnKF)can also reduce the uncertainty of soil parameters.This study combines both direct and indirect methods through multi-source data fusion to analyze soil excavation.The results indicate that combining these two methods can significantly reduce the uncertainty of soil parameters,thereby decreasing the uncertainty in structural responses.

关 键 词：空间变异性 多源数据融合 反演分析 集合卡尔曼滤波器(EnKF) Kriging法 不确定性缩减 随机场 

分 类 号：TU43[建筑科学—岩土工程]