ASPECT OF RESERVOIR CHARACTERIZATION FROM WATERFLOOD PERFORMANCE DATA

This paper presents the results of studies conducted on the behavior of producing wells in a waterflood as affected by lateral variation in rock properties, formation stratification and crossflow. The purpose of the study was to derive useful reservoir characterization information from the embedded signals in waterflood performance data. There exists a difference between the prediction of waterfloods from analytical techniques and the real performance data. Deviations of actual responses from the predicted response for an assumed homogeneous reservoir can serve to map the lateral reservoir properties.

The work on the analytics of waterflood performance has been limited. Among these, one can refer to the work by Ershaghi et al, Chan et al and Yortsos et al. In this paper we demonstrate that an in depth analysis of such observations can provide useful information in the process of reservoir characterization in terms of permeability distribution in the reservoir. The case under study here includes a waterflood under the effect the lateral variation in formation properties. We examined the behavior of producing wells in a line drive cut recovery block. We demonstrate the diagnostic capability of using delta mapping of cum oil Qo and invaded water "WI" affecting each producer. While the results presented have been tested for line drive systems, the concept can equally be applied to other well patterns.

For stratified reservoirs, we focused on the impact of a thief zone on the behavior of wells producing from sand shale sequences. Using the theoretically predicted flood performance from analytical models for such systems, as calibrated by simulation studies, we developed the metrics for the estimation of a new parameter defined here as the thief zone heterogeneity index.