Abstract

The novel Fractional Dimension RTA (FD-RTA) was applied to a recent multi-pad development campaign in the Haynesville to test the suitability of the model to represent well performance and its ability to predict production with a very limited data history calibration. Additionally, several wells exhibited parent-child effects, so FD-RTA was also applied to quantify the change in productivity in the parent wells after child wells started production. The method was applied to a total of 18 gas wells in Haynesville, all recent producers completed with modern frac designs. FD-RTA was performed on a group of 14 wells with only 3 months of data; the models were then used to predict well performance that was compared to close to 18 months of production history. Long-term predictions (50 years) were then made for each well accounting for the effect of neighboring wells on production. For parent wells, a novel methodology was applied to account for child well interference that estimates the reduction of drainage volume. Also, the paper presents data quality control tips as well as using FD-RTA as a method for initial reservoir pressure estimation. The results show a very narrow range for the parameter delta by landing zone, with an average of 0.3 in the upper target and 0.27 in the lower target. The forecast validation was successful, exhibiting an error of less than 10% in cumulative production at the end of production history for most wells. When the error was greater than 10%, the exercise was repeated using 6 months of data for calibration, which reduced the error to less than 3%. The parent well analysis resulted in excellent history matches and allowed quantifying changes in drainage area of the parent wells due to child well interaction. Initial pressure estimation described the impact of depletion on child wells. Learnings with respect to the role of natural fractures in well performance were also obtained.

This content is only available via PDF.
You can access this article if you purchase or spend a download.