A physical simulation method with a combination of dynamic displacement and imbibition was established by integrating nuclear magnetic resonance(NMR)and CT scanning.The microscopic production mechanism of tight/shale ...A physical simulation method with a combination of dynamic displacement and imbibition was established by integrating nuclear magnetic resonance(NMR)and CT scanning.The microscopic production mechanism of tight/shale oil in pore throat by dynamic imbibition and the influencing factors on the development effect of dynamic imbibition were analyzed.The dynamic seepage process of fracking-soaking-backflow-production integration was simulated,which reveals the dynamic production characteristics at different development stages and their contribution to enhancing oil recovery(EOR).The seepage of tight/shale reservoirs can be divided into three stages:strong displacement and weak imbibition as oil produced rapidly by displacement from macropores and fractures,weak displacement and strong imbibition as oil produced slowly by reverse imbibition from small pores,and weak displacement and weak imbibition at dynamic equilibrium.The greater displacement pressure results in the higher displacement recovery and the lower imbibition recovery.However,if the displacement pressure is too high,the injected water is easy to break through the front and reduce the recovery degree.The higher the permeability,the greater the imbibition and displacement recovery,the shorter the time of imbibition balance,and the higher the final recovery.The fractures can effectively increase the imbibition contact area between matrix and water,reduce the oil-water seepage resistance,promote the oil-water displacement between matrix and fracture,and improve the oil displacement rate and recovery of the matrix.The soaking after fracturing is beneficial to the imbibition replacement and energy storage of the fluid;also,the effective use of the carrying of the backflow fluid and the displacement in the mining stage is the key to enhancing oil recovery.展开更多
Current data-driven deep learning(DL)methods typically reconstruct subsurface velocity models directly from pre-stack seismic records.However,these purely data-driven methods are often less robust and produce results ...Current data-driven deep learning(DL)methods typically reconstruct subsurface velocity models directly from pre-stack seismic records.However,these purely data-driven methods are often less robust and produce results that are less physically interpretative.Here,the authors propose a new method that uses migration images as input,combined with convolutional neural networks to construct high-resolution velocity models.Compared to directly using pre-stack seismic records as input,the nonlinearity between migration images and velocity models is significantly reduced.Additionally,the advantage of using migration images lies in its ability to more comprehensively capture the reflective properties of the subsurface medium,including amplitude and phase information,thereby to provide richer physical information in guiding the reconstruction of the velocity model.This approach not only improves the accuracy and resolution of the reconstructed velocity models,but also enhances the physical interpretability and robustness.Numerical experiments on synthetic data show that the proposed method has superior reconstruction performance and strong generalization capability when dealing with complex geological structures,and shows great potential in providing efficient solutions for the task of reconstructing high-wavenumber components.展开更多
基金Supported by the PetroChina Science and Technology Major Project(2021-117)PetroChina CCUS Major Science and Technology Project(2021ZZ01-03)。
文摘A physical simulation method with a combination of dynamic displacement and imbibition was established by integrating nuclear magnetic resonance(NMR)and CT scanning.The microscopic production mechanism of tight/shale oil in pore throat by dynamic imbibition and the influencing factors on the development effect of dynamic imbibition were analyzed.The dynamic seepage process of fracking-soaking-backflow-production integration was simulated,which reveals the dynamic production characteristics at different development stages and their contribution to enhancing oil recovery(EOR).The seepage of tight/shale reservoirs can be divided into three stages:strong displacement and weak imbibition as oil produced rapidly by displacement from macropores and fractures,weak displacement and strong imbibition as oil produced slowly by reverse imbibition from small pores,and weak displacement and weak imbibition at dynamic equilibrium.The greater displacement pressure results in the higher displacement recovery and the lower imbibition recovery.However,if the displacement pressure is too high,the injected water is easy to break through the front and reduce the recovery degree.The higher the permeability,the greater the imbibition and displacement recovery,the shorter the time of imbibition balance,and the higher the final recovery.The fractures can effectively increase the imbibition contact area between matrix and water,reduce the oil-water seepage resistance,promote the oil-water displacement between matrix and fracture,and improve the oil displacement rate and recovery of the matrix.The soaking after fracturing is beneficial to the imbibition replacement and energy storage of the fluid;also,the effective use of the carrying of the backflow fluid and the displacement in the mining stage is the key to enhancing oil recovery.
文摘Current data-driven deep learning(DL)methods typically reconstruct subsurface velocity models directly from pre-stack seismic records.However,these purely data-driven methods are often less robust and produce results that are less physically interpretative.Here,the authors propose a new method that uses migration images as input,combined with convolutional neural networks to construct high-resolution velocity models.Compared to directly using pre-stack seismic records as input,the nonlinearity between migration images and velocity models is significantly reduced.Additionally,the advantage of using migration images lies in its ability to more comprehensively capture the reflective properties of the subsurface medium,including amplitude and phase information,thereby to provide richer physical information in guiding the reconstruction of the velocity model.This approach not only improves the accuracy and resolution of the reconstructed velocity models,but also enhances the physical interpretability and robustness.Numerical experiments on synthetic data show that the proposed method has superior reconstruction performance and strong generalization capability when dealing with complex geological structures,and shows great potential in providing efficient solutions for the task of reconstructing high-wavenumber components.
