Abstract
The present paper proposes a methodology to simulate samples of two-phase random media based on their first and second order moments. The proposed approach makes use of the zero-crossings of a Gaussian random field to generate a binary random field. The latter takes only two values (0 and 1) and is used to describe the random medium under investigation. An elaborate algorithm is presented that calculates the optimum autocorrelation of the underlying Gaussian field, given the autocorrelation of the binary field. Examples in one and two dimensions are provided.
| Original language | English |
|---|---|
| Title of host publication | Computational Stochastic Mechanics |
| Subtitle of host publication | Proceedings of the fourth International Conference on Computational Stochastic Mechanics, Corfu, Greece, June 9-12, 2002 |
| Editors | P D Spanos, G Deodatis |
| Publisher | Millpress |
| Pages | 325-333 |
| Number of pages | 9 |
| ISBN (Print) | 90-77017-74-7, 9789077017746 |
| Publication status | Published - 2003 |
| Event | 4th International Conference on Computational Stochastic Mechanics - Corfu, Greece Duration: 9 Jun 2002 → 12 Jun 2002 |
Conference
| Conference | 4th International Conference on Computational Stochastic Mechanics |
|---|---|
| Country/Territory | Greece |
| City | Corfu |
| Period | 9/06/02 → 12/06/02 |
Keywords
- two-phase materials
- random media
- non-Gaussian fields
- zero-crossings
- RECONSTRUCTING RANDOM-MEDIA