TY - JOUR
T1 - Machine learning approaches to predict the micromechanical properties of cementitious hydration phases from microstructural chemical maps
AU - Ford, Emily
AU - Kailas, Shankar
AU - Maneparambil, Kailasnath
AU - Neithalath, Narayanan
N1 - Publisher Copyright:
© 2020 Elsevier Ltd
PY - 2020/12/30
Y1 - 2020/12/30
N2 - This paper demonstrates the use of normalized intensities of chemical species obtained from energy-dispersive X-ray spectroscopy (EDS) as inputs to machine learning (ML) models, in order to predict the nanoindentation moduli (M) of different phases in a cementitious matrix. Single and multi-component blends belonging to conventional and ultra-high performance (UHP) pastes are evaluated using a variety of ML models. It is shown that the relative intensities of Ca, Si, and Al can be used to accurately predict the phase moduli in well-hydrated pastes with limited microstructural complexities, using all the ML models investigated. When data sets belonging to multiple binders or those for UHP pastes consisting of multiple materials and low degrees of reaction are considered, the accuracy of ML predictions are found to be significantly lower. This is partly attributable to the presence of mixed phases with widely differing chemistry-property relationships, and the lack of data for higher stiffness phases that exaggerate the skew-sensitivity of ML models like ANN. Potential data augmentation strategies to tide over some of these effects are suggested.
AB - This paper demonstrates the use of normalized intensities of chemical species obtained from energy-dispersive X-ray spectroscopy (EDS) as inputs to machine learning (ML) models, in order to predict the nanoindentation moduli (M) of different phases in a cementitious matrix. Single and multi-component blends belonging to conventional and ultra-high performance (UHP) pastes are evaluated using a variety of ML models. It is shown that the relative intensities of Ca, Si, and Al can be used to accurately predict the phase moduli in well-hydrated pastes with limited microstructural complexities, using all the ML models investigated. When data sets belonging to multiple binders or those for UHP pastes consisting of multiple materials and low degrees of reaction are considered, the accuracy of ML predictions are found to be significantly lower. This is partly attributable to the presence of mixed phases with widely differing chemistry-property relationships, and the lack of data for higher stiffness phases that exaggerate the skew-sensitivity of ML models like ANN. Potential data augmentation strategies to tide over some of these effects are suggested.
KW - Cement pastes
KW - Chemical mapping
KW - Machine learning
KW - Microstructure
KW - Modulus
KW - Nanoindentation
KW - Ultra-high performance concrete
UR - http://www.scopus.com/inward/record.url?scp=85090426231&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85090426231&partnerID=8YFLogxK
U2 - 10.1016/j.conbuildmat.2020.120647
DO - 10.1016/j.conbuildmat.2020.120647
M3 - Article
AN - SCOPUS:85090426231
SN - 0950-0618
VL - 265
JO - Construction and Building Materials
JF - Construction and Building Materials
M1 - 120647
ER -