Dynamic reservoir monitoring using similarity analysis of passive source time-lapse seismic images: Application to waterflooding front monitoring in Shengli Oilfield, China

In common practice in the oil fields, the injection of water and gas into reservoirs is a crucial technique to increase production. The control of the waterflooding front in oil/gas exploitation is a matter of great concern to reservoir engineers. Monitoring the waterflooding front in oil/gas wells plays a very important role in adjusting the well network and later in production, taking advantage of the remaining oil potential and ultimately achieving great success in improving the recovery rate. For a long time, microseismic monitoring, numerical simulation, four-dimensional seismic and other methods have been widely used in waterflooding front monitoring. However, reconciling their reliability and cost poses a significant challenge. In order to achieve real-time, reliable and cost-effective monitoring, we propose an innovative method for waterflooding front monitoring through the similarity analysis of passive source time-lapse seismic images. Typically, passive source seismic data collected from oil fields have extremely low signal-to-noise ratio (SNR), which poses a serious problem for obtaining structural images. The proposed method aims to visualize and analyze underground changes by highlighting time-lapse images and provide a strategy for underground monitoring using long-term passive source data under low SNR conditions. First, we verify the feasibility of the proposed method by designing a theoretical model. Then, we conduct an analysis of the correlation coefficient (similarity) on the passive source time-lapse seismic imaging results to enhance the image differences and identify the simulated waterflooding fronts. Finally, the proposed method is applied to the actual waterflooding front monitoring tasks in Shengli Oilfield, China. The research findings indicate that the monitoring results are consistent with the actual development conditions, which in turn demonstrates that the proposed method has great potential for practical application and is very suitable for monitoring common development tasks in oil fields.