Estimating Model Parameters of Rockfill Materials Using Neural Network Approach

 

Shen Yu

State Key Lab. of Structural Analysis for Industrial Equipment, Dalian University of Technology, Dalian 116024, China
e-mail: yushen@dlut.edu.cn

Shouju Li

State Key Lab. of Structural Analysis for Industrial Equipment, Dalian University of Technology, Dalian 116024, China
e-mail: lishouju@dlut.edu.cn

 

 ABSTRACT

In order to deal with the ill-posed problem of material parameter identification for rockfill materials, a procedure based on neural network is proposed. Duncan-Chang nonlinear constitutive model is adopted to characterize the behavior of the modeled rockfill materials. An analytical solution for computing the relationship between strain and vertical load is approached for simulating triaxial compression experiments. Based on neural network, the material parameters are determined from the triaxial compression experimental results. The investigation results show that the predicted strains provide satisfactory precision by comparing with observed ones.

Keywords: parameter identification; neural network; rockfill materials; nonlinear constitutive model; experimental investigation

Get the entire paper (pdf)    Go back to the TOC