Sedimentary Rock Reconstruction And Pore-Scale NMR Response Simulation

Nuclear Magnetic Resonance (NMR) logging is commonly used for evaluating rock properties, such as porosity, pore radius distribution and so on. If macro- and micro-pore connect, may causes pore-pore diffusive coupling phenomenon which leads to the NMR measurement cannot reflects the true pore radius distribution. So, it is necessary to thorough research the pore-scale NMR response of rock.This thesis includes two parts:sedimentary rock reconstruction and pore-scale NMR response simulation. We simulated the diagenesis processes of sedimentary rock, such as sedimentation, compaction, and cementation. The NMR responses of rock of different diagenesis process are simulated by random-walk method, and T2 distributions are achieved from BRD inversion algorithm. The Tt distribution of rock of different diagenesis process are analyzed, it is found that the porosities and pore radius are gradually decreases, which cause NMR relaxation rate accelerating and T2 distribution shift to short T2. The compaction degree influence on the rock pore structure and NMR response are compared, it is indicate that the greater strain of the compaction process, the smaller porosity and pore radius which causes the NMR surface relaxation rate increases. The rock pore structures and NMR responses from two different process-based methods are compared, it is indicate that the difference of rock pore structures and NMR relaxation signals from compaction are small, but the difference from cementation are huge. Finally, this thesis simulated the NMR response of actual sandstone, such as high porosity and permeability Berea sandstones and low porosity and permeability tight sandstone, and compared to measurements. The experimental results indicate that CT scan can distinguish the macro-porous which make the simulation long relaxation time components are preferably coincident with measurements, but cannot preferred reflect the micro-porous information which cause the huge difference between the simulation short relaxation time components and measurements.