--> --> Abstract: Petrophysical Studies in the Characterization of Carbonate Reservoirs of Campos Basin – Brazil, by A. Carrasquilla, G. Nocchi, V. Briones, M. Torres, N. Franco Filho, F. Schuab, and P. Sanchez; #120034 (2012)

Datapages, Inc.Print this page

Petrophysical Studies in the Characterization of Carbonate Reservoirs of Campos Basin – Brazil

A. Carrasquilla¹, G. Nocchi¹, V. Briones¹, M. Torres², N. Franco Filho³, F. Schuab³, and P. Sanchez4
¹UENF, Macaé, RJ, Brazil
²CENPES – Petrobras, Rio de Janeiro, Brazil
³UOBC – Petrobras, Macaé, RJ, Brazil
4Paradigm, Rio de Janeiro, Brazil

The results of basic petrophysical laboratory tests, jointly with the interpretation of Conventional (gamma rays, resistivity, density, neutron porosity and sonic), Resistivity Images (RI) and Nuclear Magnetic Resonance (NMR) Logs, were all analyzed with the objective to characterize a carbonate reservoir in Campos Basin - Brazil. In this form, the capillary pressure tests performed by mercury injection in samples were used to calibrate the T2 distribution obtained from NMR logs, which were run in two wells across the reservoir. This calibration allowed to obtain a pseudo capillary pressure curve from T2 distribution of NMR logs, increasing efficiency in the determination of the fluids trapped by capillary and the mobile fluids and permitting a better knowledge of porosity, irreducible water saturation, fluid zones and permeability curves of this reservoir. In the continuity of the work, the proposed workflow incorporated texture information arising from the processing of RI log, coupled to pore size distribution provided by inversion of T2 relaxation time spectrum of NMR log, which allowed the determination of the permeability and the electrofacies. Based on these petrophysical characteristics, we concluded that the depositional model of the study area corresponds to a carbonate platform represented by high-energy banks with shallow upward cycles, where the predominant facies ranging from oolitic grainstones to peloids and cemented packstones. The reference well was divided in these three areas, and, basis on this, we performed a normalization of the RI with the aim of identifying textural facies. Hereafter, the results were compared with a sedimentological description, allowing a control of the model, where it was observed that the resulting textures can, in general, be used to identify lithology changes in the reservoir. Then, through of a cluster analysis using the T2 distribution of NMR log, it was estimated the petrophysical facies, which reflects the particle size changes as a function of the porosity. This methodology also allowed us to group different data sets to estimate supervised electrofacies models by the classification of core facies. The obtained models from these facies give us information about texture and pore size distribution, which were used as secondary variables to refine the permeability models of the reservoir by means of neural networks. On the other hand, still today the NMR log is expensive and therefore is not performed on all wells, and, moreover, it does not exist in old wells. For these reasons, it is interesting to simulate the profile NMR-derived parameters, such as porosity and permeability, in wells where it did not exist. To achieve these objectives, we used Conventional Logs and Artificial Intelligence Techniques, as neural networks and fuzzy logic, both optimized by a genetic algorithm. The results obtained show that it is easy to simulate the porosity, but that greater care is needed in time to simulate the permeability. In all simulated cases, the performance of neural network was better. Finally, the obtained results in this work provide, therefore, important information on the characterization of this carbonate reservoir, which aim to reduce the uncertainties in the calculation of "in place" volume

 

AAPG Search and Discovery Article #120034©2012 AAPG Hedberg Conference Fundamental Controls on Flow in Carbonates, Saint-Cyr Sur Mer, Provence, France, July 8-13, 2012