Resumo

VAZQUEZ-GUILLEN, Felipe  e  AUVINET-GUICHARD, Gabriel. Simulation of Random Fields with Non Multi-Gaussian Dependence Using Copulas. Ing. invest. y tecnol. [online]. 2014, vol.15, n.4, pp.539-548. ISSN 2594-0732.

Random fields are models commonly used in civil engineering to describe spatial variability of material properties. A copula model is used in this paper to simulate random fields with non multi-Gaussian dependence. Simulations of several random fields are conducted following the proposed methodology and bivariate empirical copulas are then examined. Satisfactory results are obtained in terms of the reproduction of the corresponding theoretical copulas. By means of such simulator, it is possible to incorporate a higher degree of realism in the spatial variability model, for example in seepage flow problems wherein the presence of higher permeable continuous zones control the hydraulic behavior of the soil mass or in slope stability problems where the extension of failure surfaces may be controlled by the presence of weaker continuous zones, to mention only a few.

Palavras-chave : simulation; random fields; copulas; spatial variability; asymmetric dependence; V-transformed copula.