Rockhead profile simulation using an improved generation method of conditional random field

• DOI: 10.1016/j.jrmge.2021.09.007
• Publication Date: 2021-11-14T05:23:49Z
• Authors: Liang Han; Lin Wang; Wengang Zhang; Boming Geng; Shang Li
• Keywords: Borehole; Rockhead profile; 0211 other engineering and technologies; Mean uncertainty; TA703-712; Engineering geology. Rock mechanics. Soil mechanics. Underground construction; 02 engineering and technology; Conditional random field (CRF); Bayesian

Rockhead profile is an important part of geological profiles and can have significant impacts on some geotechnical engineering practice, thus, it necessary to establish a useful method reverse the rockhead using site investigation results. As general reflect spatial distribution geo-material properties based field measurements, conditional random (CRF) was improved in this paper simulate profiles. Besides, measurements are generally limited due limitations budget time so that estimation mean value uncertainty extent. Bayesian theory effectively combine prior information deal with uncertainty, CRF implemented aid framework study. More importantly, simulation procedure achieved as analytical solution avoid time-consuming sampling work. The results show proposed provide reasonable about depth at various locations against measurement data result, subjectivity determining be minimized. Finally, both selection hyper-parameters affect simulated profiles, while influence latter less than former.