Machine Learning Clustering of Reservoir Heterogeneity with Petrophysicaland Production Data

Reservoir development decisions strongly depend on our understanding on reservoir heterogeneity, which isoften subject to sparse and conflicting data, interpretational bias and constraints imposed by the modellingassumptions. The work tackles a challenging task of accurately and quickly identifying and describinguncertainty in the spatial distribution of reservoir heterogeneity derived from geological well data and withrespect to a geological concept. We propose a metric based machine-learning approach to identify anddescribe spatial trends in reservoir heterogeneity/facies property distribution using wireline and productiondata. We demonstrate how the proposed method can help to partition reservoir heterogeneity and discover andverify spatial trends for a real mature producing field in the Western Siberia. The obtained clustering ofreservoir facies based on the wireline logs (alpha-SP) demonstrated a good agreement with the reservoirzonation based on manual log interpretation and the geological concept. Clustering based on individualwell production profiles has confirmed the reservoir partitioning and matched some of the reservoir featuresaligned with the prevailing geological concept. The outcome of the proposed method helps to improve thefacies distribution model by integrating the discovered spatial trends into a geostatistical model and accountfor uncertainty in the depositional scenario that is difficult to quantify based on manual interpretation.