No abstract

No abstract

No abstract

Reservoir description and characterization is one of the main/critical engineering components which require a good understanding to ensure the optimum reservoir development that leads to the highest recovery. Reservoir modeling uses all available information which includes at a minimum logs data, and fluid and rock properties. In this study, a deep learning neural network was developed to estimate the petrophysical characteristics required building a full field earth model for a large reservoir. This was accomplished through a learning process whereby the model was presented with diverse and large volumes of log data measured in the field. The study demonstrates the capability of the deep learning neural network model when tested against the newly drilled wells in the field. The model proved to generate synthetic logs almost identical to those recorded in the new wells. In a future paper, we will demonstrate how the reservoir model constructed using generated data led to significant improvement in the full field reservoir as contrasted to the existing earth model developed using Kriging technology.