Evaluation of Production Quality from Molecular Signatures of Organic Matter and Mineralogy in Shale Rocks
Abstract
We will report our recent progress on molecular-level classification of organic matters and minerals using infrared spectroscopy and infrared micro-ATR imaging on samples prepared from two different types of shale rocks, one from a natural gas producing area and the other from an oil producing area. We distinguish the shales samples from different origins and depths with principal component analysis and k-means clustering on the spectra of the bulk form, and then we use deep neural network modeling to correlate the bulk properties to the microscale chemical images to improve our interpretation, especially on the organic matter and the accurate estimation of its concentration and constituents. This molecular-level understanding further provides insights on the production quality of the produced fluids and predictions of the output rate of the underlying wells.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2018
- Bibcode:
- 2018AGUFM.S14A..05K
- Keywords:
-
- 0555 Neural networks;
- fuzzy logic;
- machine learning;
- COMPUTATIONAL GEOPHYSICSDE: 1910 Data assimilation;
- integration and fusion;
- INFORMATICSDE: 1942 Machine learning;
- INFORMATICSDE: 7223 Earthquake interaction;
- forecasting;
- and prediction;
- SEISMOLOGY