Abstract

Elasto-plastic constitutive models implemented in a finite element code is a widely used tool to assess the sand production potential of unconsolidated formations. An example of such a model is N. Morita's kinematic hardening model, which has been used by several companies since its publication in 1980. This model has been a key element in Statoil's sand prediction strategy for more than a decade.

The estimation of the parameters of the model is performed by matching a set of triaxial test data as closely as possible. Due to the complexity of the model, this has been a tedious task, involving a lot of trial and error in matching the predicted stress strain curves to the laboratory data. As a result, even experienced engineers could be expected to come up with significantly different parameterisations of the model for the same set of laboratory data.

In this paper we formulate the model in a way which allows direct estimation of the model parameters from the laboratory data, eliminating much of the subjectivity in the parameter estimation. This procedure allow us to give a more precise evaluation of the strengths and short comings of the model. Examples of how well North Sea weak sandstones can be approximated by the Morita model are given.

By the reformulation of the model, we have cut the time required to set up a parameterisation drastically, and the final model is to a smaller degree dependent on the skills of the engineer. The model is found to be able to match the axial stress- axial strain curves for most datasets with acceptable accuracy. However, due to the associated flow rule of the model, it is in general not possible to get a simultaneous good fit to the radial strain data.

P. 277

This content is only available via PDF.
You can access this article if you purchase or spend a download.