Computational Intelligence for Deepwater Reservoir Depositional Environments Interpretation
This work addresses the problem of reducing workload and improving consistency for stratigraphers in oil recovery prediction, though it is incremental as it automates an existing process with new techniques.
The researchers tackled the labor-intensive and inconsistent manual interpretation of deepwater reservoir stratigraphic components by developing an automated methodology using computational intelligence techniques, demonstrating that it can produce adequate finite state transducer models from well log data.
Predicting oil recovery efficiency of a deepwater reservoir is a challenging task. One approach to characterize a deepwater reservoir and to predict its producibility is by analyzing its depositional information. This research proposes a deposition-based stratigraphic interpretation framework for deepwater reservoir characterization. In this framework, one critical task is the identification and labeling of the stratigraphic components in the reservoir, according to their depositional environments. This interpretation process is labor intensive and can produce different results depending on the stratigrapher who performs the analysis. To relieve stratigrapher's workload and to produce more consistent results, we have developed a novel methodology to automate this process using various computational intelligence techniques. Using a well log data set, we demonstrate that the developed methodology and the designed workflow can produce finite state transducer models that interpret deepwater reservoir depositional environments adequately.