MOURA, C. H. B.; http://lattes.cnpq.br/5553835796393157; MOURA, Carlos Henrickson Barbalho de.
Resumo:
Computational petrophysics is a technique that has been increasingly used in the
petroleum industry to characterize reservoirs and to simulate computationally its
physical behavior. Through this technique it is possible to characterize a big
number of samples, under different environmental conditions, in a relatively short
time. This work proposes a model of permeability estimation that uses
petrophysical parameters taken from x - ray microtomography images (µCT) and
compare them with petrophysical parameters measured in the laboratory. It was
analyzed a set of 19 samples with different depositional, diagenetic and textural
characteristics, belonging to the Araripe, Potiguar and Sergipe - Alagoas basins.
Of these, 14 are limestones, 2 of tufa limestone, 2 of caliche and 1 of dolomite.
In the laboratory a gas permoporosimeter was used to measure the porosity and
permeability parameters. µCT samples were obtained with a resolution of about
2.0 μm. The set of images created was treated in Avizo Fire software and the
porosity, permeability, connectivity and pore diameter parameters were
extracted. A statistical model was established to predict permeability from pore
space parameters extracted from µCT images. The results indicate that the
connectivity of micropores, inferred from the calculation of the Euler Number in
3D images, is the parameter that exerts the greatest influence in the estimation
of permeability, followed by the porosity of the macropores and the connectivity
of the macropores. The proposed predictive model presented a coefficient of
determination of 0.994, being very reliable for the group of samples investigated.
