Implementation of numerical optimization techniques in constitutive model calibration
Journal
Springer Series in Geomechanics and Geoengineering
ISBN
9783642328138
Date Issued
2013-01-01
Author(s)
Abstract
When a conventional method such as linear regression is insufficient to identify the constants and parameters of an advanced constitutive model, numerical optimization techniques are often used. In employing numerical optimization to constitutive model calibration, an objective function essentially needs to be formulated in the form of nonlinear least squares where the sum of the differences between the measured and computed data points are quantified. When a minimum value of the objection function is obtained, the corresponding variables are the optimized model constants and parameters. Due to the complexity in the format of objective functions in constitutive model calibration, gradient-based methods are seldom used. In this paper, non-gradient based methods namely Direct algorithm was implemented to calibrate a modified Cam-Clay model against laboratory data. © Springer-Verlag Berlin Heidelberg 2013.
Type
book