The use of Nuclear Magnetic Resonance (NMR) logs for petrophysical interpretation and evaluation in the oil and gas industry has become very important in the last decades, especially in Brazil, where the use of this technology was significantly increased in the evaluation of different formations. This is mainly because the porosity estimation provided by this technology is independent of lithology and can be associated with different pore sizes.
Nevertheless, the presence of hydrocarbon inside the pore space affect the NMR response in term of time decay. This response depends of the oil viscosity because time decay is faster for oils with high viscosity. For low gradient NMR technologies, oil decay varies from high time constant for light oil to short time constant for medium and heavy oil.
This paper presents the results of applying Gamma Inversion to process the logging-while-drilling (LWD) NMR data acquired in a well drilled in the Polvo Field, Offshore Brazil, in order to evaluate fluid viscosity.
The Gamma Inversion process uses probabilistic functions to generate a T2 spectrum, which can separate signals from different fluids and allow the analysis of fluid viscosity by comparing with simulated NMR data based on reservoir and fluid properties. The evaluated fluid viscosity from NMR data is validated with information obtained from fluid sampling and analysis.