致密滩坝砂储集层孔隙分形特征、预测及应用

储层孔隙分形可以有效表征储层孔渗性能，综合反映储层孔隙结构特征及评价储层开发效果。为确定开发区块动用次序，对于东营凹陷西部区块沙四上纯下次亚段的致密滩坝砂储层，首先，利用薄片、物性及压汞等相关测试数据，计算致密滩坝砂孔喉分形维数（D）；其次，探讨分形维数与储层物性、孔隙结构参数相关性；然后，优选测井数据，建立孔喉分形维数的测井预测模型，并对东营凹陷西部区块滩坝砂分形维数平面分布进行了预测；最后，根据分形维数和油井产能相关性分析，建立了基于分形维数的储层评价标准，对研究区的致密滩坝砂储层进行了分类与评价。结果表明:当2

Fractal Characteristics,Prediction and Application for Pores in a Tight Beach-bar Sand Reservoir: A case study for Dongying Sag

Reservoir pore fractals effectively describe reservoir porosity and permeability， as well as comprehensively reflecting reservoir pore structure and evaluating the level of reservoir development. To calculate the pore fractal dimension of tight beach-bar sand reservoirs and determine the order of production in these reservoirs and increase their production rate， the tight beach-bar sands of the upper Fourth member of the Shahejie Formation were chosen as the research target. First， 12 typical samples were selected for thin section observation， physical property testing and mercury intrusion testing. Second， the fractal dimension D of the samples was calculated from the mercury intrusion test data， wetting phase saturation definition and fractal theory. The correlation between D and physical properties and pore structure parameters are shown， and a logging prediction model for tight beach bar sand was established using the following procedure. （1） The correlation between D and conventional logging curves was analyzed； four logging parameters with a correlation coefficient r>0.86 were then selected. （2） Intervals of non-beach-bar sand facies were excluded. （3） To minimize the impact of the differences in magnitude of various logging parameters， standard normalization was performed on the selected logging curves. （4） The normalized logging parameters were fitted to the D value to establish a logging prediction formula for D. （5） The value of D predicted by the formula was compared with that calculated from the mercury intrusion data to assess the prediction accuracy and effect. （6） A correlation analysis between D and oil well productivity established a reservoir evaluation criterion for tight beach-bar sand reservoirs in the western block of the Dongying Sag， and four favorable development areas were selected. It was found that the porosity of beach-bar sand reservoirs lies mainly in the range 5%?18% （average 11.8%）； permeability is mainly 0.01×10-3 to 1×10-3 μm2 （average 0.92×10-3 μm2）. The porosity of beach sand reservoirs is mainly between 3% and 16% （average 8.9%）， and permeability is mainly from 0.01×10-3 to 1×10-3 μm2 （average 0.67×10-3 μm2）. All of these values are typical of tight sandstone reservoirs. The pore throat radius mostly has a single-peak distribution in the range 0.1 to1 μm， but with different peak values for beach-bar sand （0.183?1.529 μm， average 0.8 μm） and beach sand （0.059?0.189 μm， average 0.119 μm）. The fractal dimensions of the 12 samples all lie within the range 2≤D≤3 （bar sand 2.222 4?2.531 9， average 2.441 2； beach sand 2.585 0?2.742 4， average 2.650 9）. Analysis of the correlation between D and reservoir porosity， permeability and pore structure parameters showed that D is negatively correlated with total porosity， permeability， maximum pore throat radius， median throat radius and mercury removal efficiency， and positively correlated with displacement pressure， median pressure and sorting coefficient: in other words， an increase in D indicates worsening reservoir porosity， permeability， pore throat radius， sorting and connectivity. Four log curve parameters were selected for standard normalization， fitting and establishing a fractal dimension log prediction model: the absolute value of the micro-potential and micro-gradient difference； acoustic time difference； natural gamma radiation； and resistivity. Correlation between D predicted by the proposed model and the value calculated from the mercury intrusion data reached R2=0.936， which meets the requirements of field production and scientific research. For the blocks in the Dongying Sag not using fracturing measures， the relationship between daily oil production， cumulative oil production and fractal dimension were selected to establish an evaluation criterion based on the value of D for tight beach-bar sand reservoirs. The range 2≤D≤2.35 indicates a high-quality reservoir； 2.35≤D≤2.55 indicates a general reservoir； and 2.55≤D≤3 indicates a poor reservoir. Based on the proposed model， the presence of favorable reservoirs of tight beach-bar sands was predicted in the upper fourth member of Shahejie Formation in the western block of the Dongying Sag. High-quality reservoirs are concentrated mainly in the B424 and C107?C66?B3 well areas. This study provides a theoretical basis for determining the next step in the reservoir production sequence.