The Research Of Evaluation On Thin Layer Based On Joint Inversion Of Gamma—NMR

NMR logging obtains fluid type and rock physical parameters by measuring relaxation and diffusion information of formation fluid, but its vertical resolution(VR) is low. Natural gamma logging obtains strata lithology information by detecting natural radioactivity energy and has a high VR, but it can't do the reservoir evaluation.In order to identify and evaluate thin bed layer accurately, a binding joint inversion of natural gamma and NMR logging data is proposed, combining the advantage of the both logging methods. The steps of the method are described as follows. First, based on the principle of natural gamma logging and GR curve activity, strata is divided to get the thickness information and lithology information of different formation type. Second, due to the thin bed layer thickness information, based on the NMR logging response equation at the formation boundary and the principle of PAPS pair, an echo data calibration equation for thin bed layer is derived, with the ring noise and influence of surrounded bed eliminated. Third, under the constraints derived from T2 spectrum model and lithology, based on BRD algorithm, a high resolution T2 spectrum is calculated with a minority of distribution component which are optimally selected from standard T2 spectrum.In this thesis, data correction method is simulated first. Assumed that the thickness of thin bed layer is known, echo data of the layer is calculated through calibration equation in different formation boundary models: single peak, double peaks and triple peaks. The standard T2 spectrum inverted from the corrected data is basically consistent with constructed T2 spectrum. Then the high resolution joint inversion of T2 spectrum model with single peak is simulated. Under the condition of SNR=5, compared to the standard inversion method, peak position and total porosity can be more accurate through joint inversion. Finally, according to the application on actual data from a log, compared with conventional data processing method, the binding joint inversion improves the accuracy of the identification and evaluation of thin bed layer.