A Joint Inversion Method Algorithm for T2-Pc two Dimensional NMR based on Logistic Functions

Abstract

Data processing is a key step in the analysis of nuclear magnetic resonance (NMR) experimental data, and efficient and accurate computer inversion algorithms are the core of the T2-Pc two-dimensional NMR experiment. T2-Pc two-dimensional NMR experiment is an advanced experimental method that characterizes reservoir connectivity through two dimensions: relaxation time and capillary pressure, and can obtain irreducible water saturation under different pressure differences. However, the algorithm currently used for T2-Pc two-dimensional spectrum inversion uses a mutation kernel function, resulting in low accuracy of calculation results. This paper uses the Logistic function as the two-dimensional inversion kernel function, rewrites the inversion algorithm, and obtains more accurate inversion results. Numerical simulations have proven that the T2-Pc two-dimensional map obtained by this method not only has higher resolution, but also has greater applicability in the case of low signal-to-noise ratio and a small number of centrifugal echo groups. Practice has found that the proposed method can reduce the number of centrifugations during the experiment and significantly improve the efficiency of T2-Pc two-dimensional nuclear magnetic resonance experiments.