Streamline-based Inversion Model Using Simulated Annealing Coupled with Geostatistical Perturbation Methods

Abstract

By using streamline simulation, fast and efficient forward simulations are feasible and tracer data can be effectively integrated into an inverse model. In order to incorporate tracer data into an inverse model, a given tracer breakthrough curve is conceptualized to be cumulative breakthroughs along independent streamlines. Simulated annealing is used with two different permeability perturbation methods. One is swapping the locations of permeability and the other is re-generation of permeability, which gives much better results than the former. Simulation results in this article show that the spatial distributions and transport behaviors of the identified fields by the inversion match well to those of the reference field.

Keywords:

• simulated annealing
• streamline simulation
• streamline-based inversion
• tracer data integration

Acknowledgments

This research was supported by KOSEF research project (R01-2003-000-10230-0) and conducted through the Research Institute of Engineering Science at Seoul National University, Korea.