Michigan Technological University
ANALYTICAL AND MACHINE LEARNING MODELING OF DIRECT CARBONATION OF NATURAL SILICATE MINERALS
Abstract
dc:description.abstractThe climate-altering effects of atmospheric CO2 make it necessary to develop innovative technologies that can remove it from the atmosphere. Mineral carbonation is a scalable technology to sequestrate and store CO2 in a stable carbonate mineral form. The efficiency of the mineral carbonation process depends on both feed mineralogy and process parameters. Comprehensive modeling and prediction of mineral carbonation efficiency for different CO2 – reactive rocks has not been achieved in the past. In this thesis work, analytical and machine learning models are developed to predict the carbonation efficiency of dunite reactions. It is found that random forest can accurately predict the carbonation efficiency of these reactions with a RMSE of 4.05. These models are then applied to predict the carbonation efficiency of mine-tailings reactions. This analysis shows that an ensemble of gradient boost and decision tree accurately predicts carbonation efficiency with a RMSE of 2.45.
Degree
thesis:*- Name thesis:degree_name
- Master of Science in Chemical Engineering (MS)
- Level thesis:degree_level
- Open Access Master's Thesis
- Discipline thesis:degree_discipline
- Department of Chemical Engineering
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Hanson, William
- Committee members dc:contributor.committeemember
-
- Kaiwu Huang
- Robert Handler
- Contributors dc:contributor
-
- Lei Pan
Subjects
dc:subject × 1Identifiers
dc:identifier.*- Identifier
- https://digitalcommons.mtu.edu/etdr/1833