Pattern-based methodology for characterization of layered geological formations

• Published in: Journal of petroleum science & engineering Vol. 13; no. 1; pp. 47 - 56
• Main Authors: Katz, S., Chilingarian, G.V., Islam, M.R.
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 01.04.1995; Elsevier Science
• Subjects: Applied geophysics; Earth sciences; Earth, ocean, space; Exact sciences and technology; Internal geophysics; algorithms; models; iterative methods; reflection methods; spectra; amplitude; reflection; mathematical methods; new methods; methods; layered materials
• ISSN: 0920-4105; 1873-4715
• DOI: 10.1016/0920-4105(94)00062-9

A new pattern-driven method for characterizing layered geologic formations is presented is this paper. The method, based on the inversion of geophysical data, is tested on normal incidence reflection seismology data. The key elements of these methods are: the formation of a data-set of trial patterns with corresponding sets of geophysical models, the calculation of the values of a cost function over a data-set of trial patterns, and the identification of a subset of patterns for which a value of the cost function is close to that of its global minimum. The cost function characterizes similarity between recorded pattern and the elements in the pattern data-set. An estimation of the model parameters corresponding to the recorded pattern is performed using a set of a priori constraints on the estimated parameters and similarity values for the elements of the pattern data-set. The pattern-driven methodology has been applied to inversion of wave-forms and amplitude spectra of normal incidence-reflected waves. It is shown that in the case of a layered formation, it is possible to estimate the thickness of the layer, as well as the impedances within the layer and the layer below, as long as acoustic properties of the layer are different from those of the surrounding medium and additional constraints on the parameters of the layer are available.