【摘要】 目前低渗透储层测井评价结果与试油效益并不令人满意 ,其原因是低渗透储集层类型的差异导致其导电性和测井响应特征不同 ,而测井解释大多数还是沿用中高孔渗油气藏的解释方法。因此 ,低渗透储集层的测井解释模型要针对不同的储集层类型建立模型 ,其核心问题是利用测井资料进行低渗透储集层类型的识别。文章首先应用毛细管压力曲线、岩电实验数据、物性分析数据和相对渗透率曲线等实验资料 ,结合实际测井和测试资料 ,采用地质储层分类标准和储层的分形特征对塔中志留系储层进行分类研究 ,建立了适应井区的测井分类标准 ;然后 ,采用BP神经网络方法建立了识别储层类型的测井分类模型 ;最后应用该方法处理该层系低孔低渗储层的实际测井资料。其处理解释结果与岩心分析结论比较 ,符合率高达 91.89% ,试油结果也验证了该方法的实用性、准确性。更多还原

【Abstract】 According to the log evaluation conclu sion, the production test result of low permeability reservoir is unsatisfactory at present. The reason is that the conducti-bility and log response characteristics of low permeability reservoirs with different types are various, however the methods of interpreting the medium-high porosity and per meability reservoir were still continuously used for most of the log interpretations. Therefore the log interpretation model should be set up in consideration of the different reser voir types and its key problem is to identify the low permea bility reservoir type by use of log data. By applying the ex perimental data on capillary pressure curves, lithological-e-lectrical experiments, petrophysical analyses and relative permeability curves, etc, in combination with the actual log ging and testing data and according to the reservoir classifi cation standard and fractal character, the Silurian reservoir classification of Tazhong region was studied and a log classi fication standard suitable to the region was set up; then a log classification model of identifying reservoir type was estab lished by use of BP neural network method; and then the ac tual log data of the Silurian low porosity and low permeabili ty reservoirs were processed by applying the model. As com pared with the core analysis data, the coincidence rate of the interpretation results was up to 91. 89%. The practicability and accuracy of the model were also verified by production test results.更多还原