Utilizing Probabilistic Neural Network for Reservoir Assessment: A Case Study from an Onshore Niger Delta Field

Accurate reservoir characterization is often useful in defining the boundaries of prospective oil or gas fields as well as in lessening the likelihood of drilling ineffective wells. Identifying fluid and lithology in heterogeneous and complex reservoirs, such as those found in the Niger Delta, is a difficult task. However, conventional seismic attributes cannot easily characterize such heterogeneous and complex reservoirs. It is critical to understand the complexities of reservoir characteristics and associated uncertainties using integrated multisource data and more robust approaches. Consequently, this piece of work is aimed at characterising the reservoir 3D water saturation volume by employing the probabilistic neural network (PNN) analysis of full-stack seismic data and water saturation logs. In this case study, two reservoirs from the onshore Niger Delta's Sigma field were set aside for characterization. A high-resolution saturation volume was predicted using well log measurement and the results produced tremendous outcomes. The predicted water saturation volume and the actual saturation logs had a 0.90 normalized correlation. The outcomes validate the ability of the workflow to precisely forecast water saturation at a greater resolution.