Machine learning applications for pore space composition

Julian is fascinated with machine learning, after completing an internship with Bluware in the Summer of 2018.  In his first application at OU, he is employing a principal component analysis (PCA) method, applied to a selected set of seismic attributes.  This processes aids in identifying a meaningful combination of attributes that may allow us to better understand and map hydrocarbons in the subsurface.  Initial projects will investigate methane and undersaturated gas accumulations. Look for his initial workflows and results at San Antonio’s AAPG May 2019 conference!