作　　者： (王开燕）; (肖增佳）; (蒋彦）; (庞思宇）; (高宇航）;

机构地区： 东北石油大学地球科学学院、非常规油气成藏与开发省部共建国家重点实验室培育基地、黑龙江省油气藏形成机理与资源评价重点实验室、黑龙江省普通高等学校油气藏形成机理与资源评价重点实验室,大庆163318

出　　处： 《地球物理学进展》 2017年第4期1665-1672,共8页

摘　　要： 常规时间域地质统计学反演方法适用于井控程度高的地区,对合成记录及层位标定要求比较高.对于开发程度较低的工区而言,由于井的密度不高,且可能存在标定效果不理想的井,这样会导致地质统计学反演的缺点被放大,影响储层砂体预测的效果.基于以上常规反演的弱点,本文通过对常规时间域内的地质统计学反演方法进行改进,形成一种新的深度域内的随机模拟反演方法,可有效消除标定误差对反演结果造成的影响.通过反演结果的对比分析,表明这种反演方法对储层的预测能力好于常规反演. Geostatistical inversion method in conventional time domain is suitable for the areas with high well control degree and makes high demands on synthetic seismogram and horizon calibration.For the low developed degree work areas,it would lead to amplifying the shortcomings of geostatistics inversion and affecting the predication of reservoir sand bodies,due to the low well density and the possibility that some wells have unsatisfactory calibration effects.Based on the weaknesses of conventional inversion mentioned above,this paper improves the geostatistical inversion method in conventional time domain and forms a new stochastic simulation inversion method in the depth domain,which can effectively eliminate the effects of calibration error on inversion results.Through the comparative analysis of inversion results,it is proved that our inversion method is better than the conventional inversion method.

关 键 词： 地质统计学 标定误差 深度域 储层预测