A semi-automatically identifying lithofacies approach from wireline logs using fuzzy clustering

Lithofacies is important for reservoir evaluation. In this work, we present a workflow to define the lithofacies from wireline logs. The workflow includes three phases: in the first phase the boundaries are automatically defined from wireline logs by using the recurrence technique; in the second phase, we extract the data set from wireline logs within the boundaries and put it in a modified fuzzy c-means clustering process. The second phase results are analysed to identify the facies. Noting that our workflow can be automated if we have an available dataset including labels to build a prediction model in the third phase. We apply our workflow to a data set in Nam Con Son basin, Vietnam. The results are comparable with core data.