Microcosmic details of pore structure are the essential factors affecting the elastic properties of tight sandstone reservoirs,while the relationships in between are still incompletely clear due to the fact that quant...Microcosmic details of pore structure are the essential factors affecting the elastic properties of tight sandstone reservoirs,while the relationships in between are still incompletely clear due to the fact that quantitative or semi-quantitative experiments are hard to achieve.Here,three sets of tight sandstone samples from the Junggar Basin are selected elaborately based on casting thin sections,XRD detection,and petro-physical measurement,and each set is characterized by a single varied microcosmic factor(pore connectedness,pore type,and grain size)of the pore structure.An ultrasonic pulse transmission technique is conducted to study the response of elastic properties to the varied microcosmic details of pore structure in the situation of different pore fluid(gas,brine,and oil)saturation and confining pressure.Observations show samples with less connectedness,inter-granular dominant pores,and smaller grain size showed greater velocities in normal conditions.Vpis more sensitive to the variations of pore type,while Vsis more sensitive to the variations of grain size.Samples with better connectedness at fluid saturation(oil or brine)show greater sensitivity to the confining pressure than those with gas saturation with a growth rate of 6.9%-11.9%,and the sensitivity is more likely controlled by connectedness.The pore types(inter-granular or intra-granular)can be distinguished by the sensitivity of velocities to the variation of pore fluid at high confining pressure(>60 MPa).The samples with small grain sizes tend to be more sensitive to the variations of confining pressure.With this knowledge,we can semi-quantitatively distinguish the complex pore structures with different fluids by the variation of elastic properties,which can help improve the precision of seismic reservoir prediction for tight sandstone reservoirs.展开更多
Tight sandstone reservoirs differ fundamentally from conventional medium to high permeability reservoirs due to their complex and heterogeneous microscopic pore structures.This complexity poses significant challenges ...Tight sandstone reservoirs differ fundamentally from conventional medium to high permeability reservoirs due to their complex and heterogeneous microscopic pore structures.This complexity poses significant challenges for accurate reservoir characterization and often results in suboptimal development performance.The specific configuration of microporosity combinations plays a decisive role in determining the storage and seepage capacities of tight sandstone reservoirs.Therefore,the precise identification of microporosity combination types is essential for improving both reservoir evaluation accuracy and development effectiveness.However,traditional computer vision models exhibit limitations in capturing fine-grained textures and spatial relationships among microscopic pores with complex morphologies,leading to inadequate generalization capabilities.To address these issues,this study proposes an enhanced Swin Transformer-based neural network architecture,termed SwinLSC(Swin Transformer with Linformer and Self-Adaptive Channel Attention).The model incorporates a globallocal attention mechanism and is trained on image datasets of cast thin sections from tight sandstone reservoirs in the Yanchang Oilfield.To evaluate model performance,Top-1 Accuracy,Loss,and Recall metrics were employed,and the SwinLSC model was benchmarked against three mainstream architectures:Swin Transformer,Vision Transformer(ViT),and ResNet.Experimental results demonstrate that SwinLSC achieves a prediction accuracy of 93.3 %,significantly outperforming the comparative models.These findings indicate that the SwinLSC model effective ly addresses the generalization deficiencies of conventional approaches in recognizing microstructural features in cast thin section imagery.Consequently,it offers a robust and accurate solution for microporosity type identification,thereby providing reliable technical support for the efficient exploration and development of tight sandstone reservoirs.展开更多
基金supported by the Open Fund(PLC2020002,PLC20190507)of the State Key Laboratory of Oil and Gas Reservoir Geology and Exploitation(Chengdu University of Technology)National Natural Science Foundation of China(42004112,42274175,42030812,41974160)+1 种基金sponsored by Special projects of local science and technology development guided by the central government in Sichuan(2021ZYD0030)Natural Science Foundation of Sichuan Province(23NSFSC5311)
文摘Microcosmic details of pore structure are the essential factors affecting the elastic properties of tight sandstone reservoirs,while the relationships in between are still incompletely clear due to the fact that quantitative or semi-quantitative experiments are hard to achieve.Here,three sets of tight sandstone samples from the Junggar Basin are selected elaborately based on casting thin sections,XRD detection,and petro-physical measurement,and each set is characterized by a single varied microcosmic factor(pore connectedness,pore type,and grain size)of the pore structure.An ultrasonic pulse transmission technique is conducted to study the response of elastic properties to the varied microcosmic details of pore structure in the situation of different pore fluid(gas,brine,and oil)saturation and confining pressure.Observations show samples with less connectedness,inter-granular dominant pores,and smaller grain size showed greater velocities in normal conditions.Vpis more sensitive to the variations of pore type,while Vsis more sensitive to the variations of grain size.Samples with better connectedness at fluid saturation(oil or brine)show greater sensitivity to the confining pressure than those with gas saturation with a growth rate of 6.9%-11.9%,and the sensitivity is more likely controlled by connectedness.The pore types(inter-granular or intra-granular)can be distinguished by the sensitivity of velocities to the variation of pore fluid at high confining pressure(>60 MPa).The samples with small grain sizes tend to be more sensitive to the variations of confining pressure.With this knowledge,we can semi-quantitatively distinguish the complex pore structures with different fluids by the variation of elastic properties,which can help improve the precision of seismic reservoir prediction for tight sandstone reservoirs.
文摘Tight sandstone reservoirs differ fundamentally from conventional medium to high permeability reservoirs due to their complex and heterogeneous microscopic pore structures.This complexity poses significant challenges for accurate reservoir characterization and often results in suboptimal development performance.The specific configuration of microporosity combinations plays a decisive role in determining the storage and seepage capacities of tight sandstone reservoirs.Therefore,the precise identification of microporosity combination types is essential for improving both reservoir evaluation accuracy and development effectiveness.However,traditional computer vision models exhibit limitations in capturing fine-grained textures and spatial relationships among microscopic pores with complex morphologies,leading to inadequate generalization capabilities.To address these issues,this study proposes an enhanced Swin Transformer-based neural network architecture,termed SwinLSC(Swin Transformer with Linformer and Self-Adaptive Channel Attention).The model incorporates a globallocal attention mechanism and is trained on image datasets of cast thin sections from tight sandstone reservoirs in the Yanchang Oilfield.To evaluate model performance,Top-1 Accuracy,Loss,and Recall metrics were employed,and the SwinLSC model was benchmarked against three mainstream architectures:Swin Transformer,Vision Transformer(ViT),and ResNet.Experimental results demonstrate that SwinLSC achieves a prediction accuracy of 93.3 %,significantly outperforming the comparative models.These findings indicate that the SwinLSC model effective ly addresses the generalization deficiencies of conventional approaches in recognizing microstructural features in cast thin section imagery.Consequently,it offers a robust and accurate solution for microporosity type identification,thereby providing reliable technical support for the efficient exploration and development of tight sandstone reservoirs.
