Abstract: Artificial Intelligence and Mathematical Modeling Applied to Wireline Log Interpretation
Khalid Amin Khan
Wireline log interpretation is a powerful tool for evaluation of formation characteristics which in turn are used for identification of reservoir and estimation of recoverable hydrocarbons. The log data is stored and plotted using computer applications but manual procedures are mostly adopted for reading the logs and interpreting their signature into the subsurface formations.
Keeping in view the new trend of applying artificial intelligence and neural networks to geological applications an artificial intelligence based approach has been developed for computerized interpretation of logs.
The working of this technique involves mathematical models based in curve analyzers which take into account the changing trend of each log curve and then pool out parameters to mathematical equations or directly to the artificial intelligent logic. Such logic has been developed for interpreting all major types of logs like spontaneous potential, gamma ray, sonic, density, neuron and various types of resistivity logs. Thus each log is interpreted independently, for example sand line and shale base lines are automatically marked from spontaneous potential logs and clay percentage is mapped from gamma ray logs.
Finally an integrated artificial intelligent analysis takes into account all the logs, their independent interpretations and the local geological information and interprets the subsurface geological picture comprising of various formations.
AAPG Search and Discovery Article #90956©1995 AAPG International Convention and Exposition Meeting, Nice, France