基于大数据的定量措施选层技术

Quantitative stimulation layer selection technology based on big data

为减少措施选层的主观性、盲目性和风险性,提高措施改造效果,提出了基于大数据的定量措施选层技术。通过对目标区储层措施前静态参数、动态参数、措施改造参数与措施改造效果关系的统计分析,优选出影响目标区域措施改造效果的敏感因子,运用模糊数学方法建立敏感因子与措施效果之间的关系模型,采用模型计算的欧氏贴近度值判断目标井层压裂的可行性。在乌里雅斯太凹陷优选出了11项影响压裂效果的敏感因子,确定了判断是否压裂的欧式贴近度值为0.425。经太27-1X井和太67井实例验证,预测结果与实际相符率达到了100%。该技术为同类砂岩储层措施选层的可行性提供了理论依据。

In order to reduce the subjectivity,blindness and risks in selecting stimulation layers and improve the effect of stimulation,a quantitative stimulation layer selection technology based on big data was proposed.Through statistical analysis on the relationship between the static parameters,dynamic parameters,stimulation parameters and the stimulation effect in the target reservoirs,the sensitive factors affecting the stimulation effect were selected.Then,a relation model between the sensitive factors and the stimulation effect was established by using fuzzy mathematics method,and the feasibility of fracturing in the target layer was judged by using Euclidean proximity value calculated by the model.Eleven sensitive factors affecting the fracturing effect were selected in the Uliastai sag,and the Euclidean proximity value for judging whether fracturing was needed was 0.425.Field application in Well Tai 27-1X and Well Tai 67 verifies that the prediction results are 100%consistent with the actual results.This technology provides a theoretical basis for the feasibility of stimulation layer selection in similar sandstone reservoirs.