Use of PNN and Post Stack Inversion to Predict Reservoir Characterization in the Mediterranean Sea, Egypt, Sapphire Field

Abstract We face the difficulty of describing the reservoir based on the availability of diverse seismic methods in complicated geologic settings with a high degree of heterogeneity in reservoir features, such as submerged channel complexes as in the Nile Delta province. However, a lack of available wells and seismic data makes using post-stack seismic inversion procedures the best way to predicate gas accumulation. The average correlation coefficient between synthetic and seismic data is 0.997, with a 7% error, demonstrating the value of model-based inversion, furthermore, because of its separate nonlinear interaction with conventional seismic characteristics and seismic inversion products, quantitative prediction of water saturation (Sw) from seismic. Water saturation prediction away from the well is critical for effectively identifying reservoirs. As a result, probabilistic neural network (PNN) analysis has become popular. Using full-stack seismic and Sw logs, (PNN) analysis was used to forecast Sw, Vsh, and Фeff 3D volume. We used the proposed neural network approach to late Pliocene gas sandstone reservoirs, Sapphire field, in the West Delta deep marine (WDDM) concession, offshore Nile Delta, Egypt, in this case, study. The discovered volume indicates a high volume of gas and condensate on numerous channel levels.