Inverse Analysis for Identifying Soil Properties
in Excavation Analysis by Elasto-plastic Soil Spring Model
Bong Kyun Kim
So far ground behaviors, in most cases, has been predicted through site investigations. However the risk of the ground excavation cannot be estimated precisely because of both estimation lacking of the objective creterion and difficulty of the post excavation analysis using in-situ data.
In this study in order to eliminate these defects inverse analysis technique which apply in-situ data in analysis procedure is developed. This technique may provide reliability in the prediction of the structure behavior and help to carry out ground excavation more safely. Especially inverse analysis technique in this paper minimizes the errors between in-situ data and numerical results applying least square method.
Inverse analysis technique requires sensitivity of response variable to design variable, so direct differentiation method is applied due to its simplicity of calculation and applicability to hystery-dependent materials. Moreover In-situ data of response variable, elementary input data for inverse analysis, have very fewer dof's than those of analysis, so in order to obtain reliability of inverse analysis, eigenvectors of soil structure are applied to extrapolate each node with in-situ data.
In this paper, single-step inverse analysis using final excavation step data as input data as well as multi-step inverse analysis using all the excavation step data as input data is developed, and each result from them is compared.
least square method, response variable, design variable, sensitivity, direct differentiation method, eigenvector, single-step inverse analysis, multi-step inverse analysis