Comprehensive Logging Evaluation Of Tight Gas Reservoirs And Distribution Patterns Of Favorable Reservoirs In Pyroclastic Rock In Dehui Fault Depression

With the development of volcanic rock oil fields,pyroclastic rocks have been discovered to be one of the important reservoir lithology for volcanic reservoirs.As a constituent material of pyroclastic rocks,pyroclastic material is a kind of transitional lithology between lava and normal sedimentary rock.Currently,large-scale volcanic reservoirs have been discovered worldwide.The pyroclastic rock reservoirs in the third member of the Huoshiling Formation in the deep Dehui Fault Depression of the Songliao Basin are widely distributed and have great potential for natural gas resources.The reservoir lithology in the study area is complex,including three major categories and eight sub categories.Affected by diagenesis,the main types of reservoir space in the study area include pores（blowhole）,dissolution pores（devitrification pores）,matrix pores（groundmass porosities）and micro-fractures（micro-cracks）.The pore structure of reservoirs is complex,making it difficult to distinguish reservoir types.The existing conductivity models cannot describe the conductivity laws accurately,which makes it difficult to qualitatively and quantitatively identify the fluid.The distribution of gas and water in the reservoir cannot be determined precisely.The unclear influencing factors of favorable reservoirs make it difficult to predict high-quality reservoirs.Therefore,it is of great significance to conduct a comprehensive logging evaluation of pyroclastic rock tight gas reservoirs in the Dehui Fault Depression and study the distribution patterns of favorable reservoirs.Based on the lithological classification scheme obtained from diagenesis and particle size,various lithological distribution patterns of pyroclastic rocks have been proposed.The development relationship of various lithology of pyroclastic rocks between volcanic passages and lakes is presented.Pyroclastic lava,normal pyroclastic rock and sedimentary pyroclastic rock in sequence are developed from volcanic passage to lakes.Combining core observation and thin section identification,two lithology recognition methods have been established based on the classification scheme of lithology,including the combination method of intersection chart and FMI image pattern,and the support vector machine method based on decision tree node information.The latter method was optimized to achieve accurate identification of eight types of pyroclastic rock lithology in the study area with a model accuracy of 98.4%.This article designs experiments such as casting thin sections,high-pressure mercury injection,nuclear magnetic resonance and nano CT scanning of pyroclastic rock reservoirs.On this basis,the impact of diagenesis on the degree of pore development in pyroclastic rock reservoirs was analyzed,and the main types of reservoir spaces with different lithologies were identified.Pyroclastic lava mainly develops pores（blowhole）,dissolution pores（devitrification pores）and micro-cracks.Normal pyroclastic rocks are mainly developed with dissolution pores（devitrification pores）,micro-cracks and matrix pores（groundmass porosities）.Sedimentary pyroclastic rocks mainly develop matrix pores（groundmass porosities）.Based on the reservoir physical properties,pore structure,reservoir space types,and productivity characteristics,different reservoir type classification schemes have been established.The sensitive parameters for reservoir type identification were optimized and pyroclastic rock reservoir identification charts were established.The intersection chart of porosity-permeability and porosity-displacement pressure has the best recognition effect with the accuracy higher than 93.75%.For different reservoir types,equations for converting nuclear magnetic resonance T₂ spectra into pseudo capillary pressure curves were established based on large and small pores,and pore structure parameter calculation models were provided.Based on ECS data,the rock skeleton parameters of eight types of pyroclastic rock lithology were determined.Considering the influence of natural gas,the density-neutron method with variable skeleton parameters is selected as the interpretation model for the porosity of pyroclastic rocks.The flow zone indicator（FZI）was applied to classify reservoirs,and the support vector machine method based on decision tree node information was used to predict reservoir types.Permeability interpretation models were established for different reservoir types,achieving high-precision evaluation of permeability in pyroclastic rock reservoirs.Based on the rock-electric experimental data,the conductivity laws of pyroclastic rock reservoirs were studied according to different lithology and reservoir types,and the results showed that the conductivity law did not comply with the Archie equation.The pores（blowhole）,dissolution pores and devitrification pores of pyroclastic rock reservoirs have the characteristics of large pores and good connectivity,while the pores of matrix pores are small and poor connectivity.It is necessary to consider the differences in connectivity between large and small pores,the influence of pore structure on rock conductivity,and the development of gas and movable water in large pores while bound water in small pores.The four components effective medium conductivity model for pyroclastic rock reservoirs was established,considering the advantage of the effective medium conductivity theory in describing the connectivity of any component and the influence of pore structure on rock conductivity.It was found that the effective medium conductivity model can describe the conductivity law better than the equivalent rock element conductivity model and pore geometry conductivity model.Accurate determination of water saturation for different reservoir types of pyroclastic rocks has been achieved.Using the established interpretation methods for lithology,reservoir type,reservoir parameters,and fluid properties,a comprehensive logging evaluation was conducted on 9wells in the study area,and the interpretation results were consistent with the gas testing conclusions.The distribution pattern of gas and water in pyroclastic reservoirs in the study area has been clarified,with gas reservoirs mainly distributed in wells N13,N11,and N18and their vicinity;Poor gas reservoirs are mainly distributed in wells N18,N12,N14,and N19 and their vicinity;The gas and water layers are mainly distributed in wells N13,N11,and N15 and their vicinity;The gas bearing water layer is mainly distributed in the N15 and N16 well and their vicinity.The lower limit of effective reservoir porosity is 6.0%,and the lower limit of permeability is 0.001×10^-3μm²,identified Class I and Class II reservoirs as effective reservoirs in the study area.Class I reservoirs are mainly distributed in wells N11,N13,N16,and N20 and their vicinity,while Class II reservoirs are mainly distributed in wells N11,N12,N13,N15,N16,N18,N19,and N20 and their vicinity.Conducting research on the influence of factors such as lithology,diagenesis,volcanic facies,and fracture development on the development of favorable reservoirs,it was clarified that tuff lava,tuff lava,and tuff containing breccia,as well as dissolution（devitrification）,near crater and transitional phases,and the development of open fractures are favorable factors for the development of favorable reservoirs in pyroclastic rocks in the study area.Based on the distribution of Class I and II reservoirs,the distribution pattern of gas and water layers,and the development pattern of favorable reservoir factors,it is predicted that wells N11,N13,and N18 and their vicinity,are favorable reservoir development areas.The predicted results are in good agreement with the NH1 gas test results of the horizontal well,proving the accuracy of this article in predicting favorable reservoirs of pyroclastic rocks.The research results have improved the logging evaluation accuracy of pyroclastic rock tight gas reservoirs in the Dehui Fault Depression.The predicted distribution patterns of high-quality reservoirs have guiding significance for the design of well locations in the work area.This article proposes various methods applied to pyroclastic rock reservoirs,including lithology distribution patterns and identification methods,reservoir type classification schemes and discrimination methods,and four component effective medium conductivity models.These methods have important reference value for the interpretation and evaluation of similar complex pyroclastic rock reservoirs.