IDENTIFYING RESERVOIR FLUIDS IN SANDY CLAY AND CARBONATE RESERVOIR USING THE WAVELET TRANSFORM WITH WELL LOGS

摘要

The primary aim of this work is to identify with high relative resolution, existing fluids in representative formations in a gas and oil reservoir, by using the one dimensional discrete wavelet transformed (DWT- lD) to geophysical well logs data (GWL) One important application in the RGP evaluation/assessment is the determining of fluids, lithology and porosity in a reservoir, logs do contain relevant information of the micro pores structure, fluids in the pore and lithology, capable to associate with high, medium and low frequencies (scales), they make the DWT-lD mentioned before, be formally suitable for this purpose. This transform is generally applied in the conditioning, and processing of signals and images (multi-resolution analysis) Results obtained in the application of the DWT-lD transform are displayed in this work, in the fluid identification, and characterization associated to sandclay and carbonated formations. The type of fluid is identified starting from the electric induction logs data from non cased wells, in the case of cased wells, the main log is the pulsed neutron compensated (PNC). Specifically, it presents results obtained from the DWTlD application in the fluid differentiation from a formation. The method herein developed, to determine the type of fluid, consists of the variance distribution evaluation (energy) regarding the scale, its petrophysics correlation and estimation of the conditional probability (Bayes analysis). The activities performed, with the aim of increasing the certainty of the identification of several fluids and its distribution in sand-clay and carbonated formations consisted on: the obtaining for the direct representative petrophysical parameters, coefficients distribution evaluation (plain timefrequency) and associated scales (multiscale analysis) form the PNC and electric induction logs; logs reconstruction, petrophysics correlation and calibration of results. From the evaluation and analysis of the results obtained it is concluded that it is possible to perform the analysis and synthesis of the different scales involved, selectively eliminating the associated contributions with factors non related to the determining petrophysical parameters and concentrate the analysis in the representative and consistent with each of the selected scales. Finally the multiresolution analysis shows to be a suitable and useful tool in the GWL data analysis and assessment. It represents a relevant contribution in hydrocarbon indirect detection.