Application of Artificial Neural Network in seismic reservoir characterization: a case study from Offshore Nile Delta

Abstract The Prediction of the reservoir characteristics from seismic amplitude data is a main challenge. Especially in Nile Delta Basin, where subsurface geology complex and reservoirs are highly heterogeneous. Modern characterization methodologies spanning around attributes analysis, deterministic stochastic inversion methods, Amplitude Variation with Offset (AVO) interpretations, stack rotations. These proved good outcomes detecting gas sand quantifying properties. However, when pre-stack not available, most AVO-related methods cannot be implemented. Moreover, there no direct link between properties, such as hydrocarbon saturation, many assumptions imbedded results questionable. Application Artificial Neural Network (ANN) algorithms to predict new emerging trend. advantage ANN algorithm over other ability build nonlinear relationships petrophysical logs data. Hence, it can used various properties 3D space reasonable amount accuracy. We implemented method on Sequoia field, Offshore Delta, chosen was Probabilistic (PNN). One well kept apart analysis later blind quality control test results.