Probabilistic maximum safe slope angle estimation using Bayes method with prior information provided by virtual excavation

摘要

A case study of maximum safe slope angle (MSSA) based on Bayes estimation was performed and a new technique called "virtual excavation" was proposed to provide prior information. Fractures orientation data was collected from rock outdrop located at the planned project site of converter station at southeastern Dalian port. Kinematic analyses with respect to wedge failure were used to determine MSSA. Bayes estimation, an effective statistical treatment technique, was applied to optimize the MSSA. Prior information is important for the efficacy of Bayes estimation. However, lack of previous data usually causes difficulties for prior information acquirement. Therefore, "virtual excavation", which means to assume a virtual slope at the investigation area parallel to the target slope being investigated (the virtual and the target slope have the same dip direction), was proposed as an effective way to get prior information. The MSSA of the virtual slope was used as prior information and the MSSA of the target slope was used as sample information. The optimal values of MSSA were 35.9°, 41.8°, 45.2°for cut slope dip directions of 38°, 116°, 299°, respectively. The proposed techniques arguably result in an improved level of confidence in calculation of MSSA in preliminary investigation.