油藏动态运移监测时移大地电磁技术研究

Most of the oil industry is facing the challenges of late stage exploit in high water bearing reservoirs. Oil-water identification, remaining oil distribution and dynamic monitoring of reservoirs have become the major technical issues in exploit in oil and gas accumulation area. Due to the small velocity difference between water and oil, time-lapse seismics cannot indentify reservoir fluids effectively. However, the notable conductivity difference between oil and water have provided physical base for electromagnetic methods in monitoring different reservoir fluids. Traditional time-lapse magnetotelluric inversion is independent inversion that cannot remove systematic errors, resulting erratic difference resistivity models. Difference inversion technique is implemented into the inversion process of time-lapse magnetotellruic data in this study to monitor oil migrations. It effectively removes systematic errors and precisely resolves the dynamic changes of conductivities in the subsurface. Numerical simulations are firstly carried out to study the superiorities of difference inversion technique in resolving the oil-water interfaces, different inversion parameters, regularizations and physical constraints are also investigated to find the most suitable inversion scheme for the processing of time-lapse magnetotelluric data. Real field time-lapse magnetotelluric data is analyzed with difference inversion technique to obtain high resolution, high precision three- dimensional and four-dimensional resistivity models. At last we aim to obtain a time-lapse magnetotelluric inversion technique for tracing the sharp oil-gas interfaces in reservoirs and to actualize time-lapse magnetotelluric technique in monitoring reservoir dynamic migrations and predicting remain oil distribution.

油水识别、剩余油分布及油藏动态监测等已成为油藏开采中的聚焦问题。由于储层中油与水速度差异小，时移地震法不能有效地识别储层中流体性质。而原油与水电阻率的显著差异，为电磁法流体识别提供了物理基础。本课题拟研究时移大地电磁技术监测油藏动态运移。传统时移大地电磁反演采用独立反演法不能有效地消除系统误差，使最终结果存在偏差。本课题拟利用差值反演法实现储层流体精确识别。通过数值模拟验证差值反演法在复杂地质条件下追踪水驱前沿的优势，研究适用于油藏动态监测的差值反演模型约束和物性约束，获得储层高分辨率、高精度成像；利用差值反演技术指导实测时移大地电磁油藏资料的解释分析，建立油藏三维、四维电阻率成像；最终获得储层油水分界面追踪的时移大地电磁反演技术，实现时移大地电磁技术油藏储层动态监测与剩余油分布预测。

油水识别、剩余油分布及油藏动态监测等已成为油藏开采中的聚焦问题。由于储层中油与 水速度差异小，时移地震法不能有效地识别储层中流体性质。而原油与水电阻率的显著差异， 为电磁法流体识别提供了物理基础。本课题拟研究时移大地电磁技术监测油藏动态运移。传统 时移大地电磁反演采用独立反演法不能有效地消除系统误差，使最终结果存在偏差。本课题拟 利用差值反演法实现储层流体精确识别。通过数值模拟验证差值反演法在复杂地质条件下追踪 水驱前沿的优势，研究适用于油藏动态监测的差值反演模型约束和物性约束，获得储层高分辨 率、高精度成像；利用差值反演技术指导实测时移大地电磁油藏资料的解释分析，建立油藏三 维、四维电阻率成像；最终获得储层油水分界面追踪的时移大地电磁反演技术，实现时移大地 电磁技术油藏储层动态监测与剩余油分布预测。研究结果表明，二、三维时移大地电磁反演总体上能够比较清晰地刻画出各时移阶段油水界面的位置，定量地给出注入水体与油藏的空间分布区域。利用反演结果，能够有效地预测出剩余油藏的分布范围。但由于大地电磁分辨率的限制，反演结果对于垂直电性界面的刻画有所欠缺，与真实模型具有一定的差距，因此如何对垂直界面进行约束，是电磁法反演中的一个难点。此外，相比于独立反演，时移数据的差异反演通过消去时移数据中所包含的系统误差，能够更准确地刻画油藏的变化区域和变化的幅度特征，并且有助于消减反演中的虚假异常特征；在此基础上，通过带约束的差异反演，能够有效地消除反演中产生的虚假异常。实测数据的二维及三维反演结果总体上反映了油藏开采过程中油藏储层电性的变化过程，但由于缺少必要的地质及井位资料，无法做出相应的地质解释。