Waterflood projects on land often well spacing for development, e.g. 5-spot pattern with 40-acre spacing. Deep water offshore waterflood projects, in contrast, are limited to far fewer wells, and hence, well density and location are critical design parameters.

Numerical simulators are often used to optimize these parameters, frequently at considerable expense, using manual trial-and-error techniques and three-dimensional (3D) finite difference simulators.

This study applies an automated technique for the optimization of well locations in a North Sea reservoir developed by waterflooding. An optimization routine was combined with a two-dimensional (2D) areal front tracking reservoir simulator. The simulator is characterized by being very CPU efficient, independent of grid orientation, and free of numerical dispersion.

With the objective of maximizing pretax net present value of oil production, the optimizer was used to find the optimal well locations for two injectors and two to three producers. Optimization runs were performed for a seven-well development case and an alternative nine-well development case. Both cases were examined using two different flood directions.

Results from several optimization runs are presented, including the optimal well locations found. In addition, production data are compared for different optimization runs.

The procedure proved to be both CPU and man-time efficient, and is recommended as a screening method prior to using large, more detailed 3D numerical models.

You can access this article if you purchase or spend a download.