upMain HomePage
Next:Introduction

Optimization of Non-Linear Gravity Models through
Generalized Simulated Annealing

K. C. Mundim - T. J. Lemaire - A. Bassrei
Universidade Federal da Bahia, Instituto de Física
Rua Caetano Moura, 123, 40210-340 Salvador BA, Brazil

Abstract:

        In this paper we apply the Generalized Simulated Annealing (GSA) approach to the inversion of gravity data for 2-D and 3-D density distributions. We consider a modeling process where the input is the vector of model parameters tex2html_wrap_inline252 (that can be density contrast, mass or some coordinates) and the output is described by the transformation tex2html_wrap_inline254 , where tex2html_wrap_inline256 is the vector of data parameters, which generally we have access in practical problems. If the vector tex2html_wrap_inline256 describes the observed actual output of the system, the problem is to ``choose'', or estimate, the parameters tex2html_wrap_inline260 in order to minimize, in some sense, in our case in the least squares sense, the difference between the observed vector tex2html_wrap_inline256 and the prescribed output of the system tex2html_wrap_inline264 . The tests with synthetic data show the promising application of GSA in gravity inversion. The results obtained suggest us that the GSA approach enables to find quickest machines than the two conventional ones (Boltzmann and Cauchy machines).

Key-words: ill-posed problems, gravity inversion, non-linear optimization, Generalized Simulated Annealing.


e-mail: Kleber Mundim