Machine Learning Application to Permeability Prediction Using Log & Core Measurements: A Realistic Workflow Application for Reservoir Characterization

The use of Artificial Intelligence continues to grow in popularity within the geosciences in view of ever-growing complexity and magnitude of available subsurface data. This is equally evident by the need for faster and accurate interpretations required to find hydrocarbons in ever more challenging and increasingly complex basins. This drive is made necessary in a continuously evolving and cost conscious petroleum industry business environment. Advances in computing architecture now easily allows for more common application of machine learning techniques in day to day geoscience workflows. The use of machine learning in permeability prediction is becoming ever more common place as more specialists adopt this technique for modelling and prediction purposes. Typical machine learning techniques include Fuzzy Logic, Artificial Neural Networks (ANN) and Self Organizing Maps (SOM) amongst others which are run both in supervised and unsupervised modes. The described workflow in this paper was carried out using an available commercial standard petrophysical package with ANN built in modules. This paper describes a typical workflow for predicting reservoir permeability based on an integrated workflow utilizing core measurements integrated with available log data. Permeability is a key rock parameter for understanding fluid flow dynamics and flow rates and its modelling usually poses some unique challenges. Traditionally and statistically, this can be done at a fairly coarse level in cored wells by utilizing Poro-Perm correlations that usually do not capture fine scale variability observed at core scale measurement. These Poro-Perm transforms are subsequently applied on uncored wells to predict permeability. This paper analyses a workflow that aims to utilize a depth-normalized log and core data set trained using an Artificial Neural Network (ANN) module, blind tested on few key cored wells and subsequently used to predict permeability in uncored wells. In conclusion, the recommended workflow will ensure much more realistic and better matching permeability predictions.