Abstract
The following quote was obtained by the news editors from the background informa tion supplied by the inventors: “Well log interpretation may be at least partial ly implemented by software, that is, automated so that at least some of the proc essing is done by a computer. For example, well logs may be correlated (e.g., ac cording to depth) using computer-processing algorithms, including machine learni ng “To facilitate successful machine learning implementations in field-scale automa ted well log interpretation processes, the logs may be corrected for consistency within a well. Currently, the log corrections are performed manually and based on formation types that can be used as reference for having a value, such as pur e calcite, anhydrite, and halite. This process is time consuming, and calls for the presence and identification of such layers. Further, there is not a systemat ic way to provide an uncertainty analysis to be used in future applications.
NETL
NSTL