Method for gradually deforming at least part of a stochastic model of Gaussian type

摘要

The model is defined by a set of parameters relative to the structure of the medium. At least two (p) independent versions (Ri) of at least part of the model are used in a stage of gradual deformation of the model by linear combination. The coefficients ( alpha ) of this combination are such that the sum of their squares is equal to 1. A number p (at least equal to two) independent versions (Ri) of at least part of the chosen model of the medium, among all the possible versions are chosen. The method includes a number of iterative steps of gradual deformation, with at each stage a composite version being obtained using the previous step with q new independent versions pulled into the set. The model is deformed by modifying the relative statistical parameters of the structure of the medium, and whilst maintaining continuity between the parts of the model. The coefficients ( alpha ) of the linear combination are trigonometric functions. A Gaussian type model such as log normal or truncated Gaussian is used.