Processing And Interpretation Of MRIL-P Nuclear Magnetic Resonance Logging Data In Zone X7

Nuclear magnetic resonance logging(NMR)is one of the important logging methods for reservoir evaluation today.Its advantage lies in its ability to directly obtain reservoir fluid information,which solves many problems in the process of oil exploration and development.In areas with complex lithology,nuclear magnetic resonance logging is one of the few effective logging techniques,especially in terms of pore structure.It has almost become the only research method,so the study of nuclear magnetic resonance logging has always been of great significance.This article conducts research on echo data denoising methods,inversion methods,variable lateral relaxation time(T2)cutoff values,and pore structure,and evaluates them from both theoretical and practical perspectives.Firstly,the denoising of the echo data is mainly processed using the improved Empirical Mode Decomposition(CEEMDAN)method.Secondly,the inversion of echo data was based on the optimization idea to establish a T2 spectrum optimization inversion model,which was solved using the differential evolution algorithm(DE)to obtain the T2 spectrum.Then,the relationship between the T2 cutoff value of the core and the T2 spectrum morphology was explored,and the variation pattern of the T2 cutoff value on the T2 spectrum was determined to obtain the variable T2 cutoff value.Furthermore,the bound water content,shale content,and permeability were calculated;At the same time,on the basis of determining the power function conversion relationship between core capillary pressure and T2,an optimization model for calculating the optimal conversion parameters was established,which realized the continuous quantitative evaluation of pore structure.Finally,the method proposed in this paper is applied to practical logging for interpretation and evaluation.Research has shown that CEEMDAN has a good denoising effect,solving the aliasing mode problem of empirical mode decomposition,improving the signal-to-noise ratio(SNR)of the echo signal,retaining a large amount of information,and effectively recovering the real signal;The Differential evolution performs well in the inversion of echo signals with SNR ≥20.For echo signals with SNR<20,the long relaxation part of T2 spectrum retrieved by the algorithm is more accurate,while the short relaxation part is poor.However,the noise reduction processing of echo data solves this problem;The T2 cutoff value determined by mathematical morphology method and the pseudo capillary pressure curve constructed were used to calculate the bound fluid,shale content,permeability,and micro pore structure parameters,respectively.The calculation results were compared with core data and were more accurate.A set of nuclear magnetic resonance logging data processing and interpretation methods has been achieved through the above research.The parameters such as shale content,porosity,permeability,average pore throat radius,mean radius,and median pressure obtained from nuclear magnetic resonance logging data processing of wells X1,X2,and X3 in X7 area have high accuracy,indicating that this method has certain practical value.