Testing Novel Portland Cement Formulations with Carbon Nanotubes and Intrinsic Properties Revelation: Nanoindentation Analysis with Machine Learning on Microstructure Identification

Abstract

Nanoindentation was utilized as a non-destructive technique to identify Portland Cement hydration phases. Artificial Intelligence (AI) and semi-supervised Machine Learning (ML) were used for knowledge gain on the effect of carbon nanotubes to nanomechanics in novel cement formulations. Data labelling is performed with unsupervised ML with k-means clustering. Supervised ML classification is used in order to predict the hydration products composition and 97.6% accuracy was achieved. Analysis included multiple nanoindentation raw data variables, and required less time to execute than conventional single component probability density analysis (PDA). Also, PDA was less informative than ML regarding information exchange and re-usability of input in design predictions. In principle, ML is the appropriate science for predictive modeling, such as cement phase identification and facilitates the acquisition of precise results. This study introduces unbiased structure-property relations with ML to monitor cement durability based on cement phases nanomechanics compared to PDA, which offers a solution based on local optima of a multidimensional space solution. Evaluation of nanomaterials inclusion in composite reinforcement using semi-supervised ML was proved feasible. This methodology is expected to contribute to design informatics due to the high prediction metrics, which holds promise for the transfer learning potential of these models for studying other novel cement formulations.

Mechanical phase mapping of the Taza meteorite using correlated high‐speed nanoindentation and EDX

“…With the advent of modern, high speed nanoindentation techniques [12,13,14], it is now possible to map the mechanical features of materials over square millimeters of area with micron-scale resolution in a reasonable amount of time. Using statistical analysis [15] or machine learning [16,17], the properties of each of these phases can be extracted. This has been applied on a range of materials from cemented carbides [13,18,19], steels [20], and thermal barrier coatings [21] to actual cements [22,23].…”

Influence of indentation size and spacing on statistical phase analysis via high-speed nanoindentation mapping of metal alloys

“…Primarily, this has been conducted using the statistical property analysis technique proposed by Ulm et al [4,[17][18][19]. However, some additional techniques have also recently been applied: K-means cluster analysis [16] and machine learning [20,21]. Both of 2 Invited Paper these techniques also allow the spatial positions of the indents to be considered as an additional dimension during the analysis allowing visualizing of the clusters within the maps.…”

On Phase Identification of Hardened Cement Pastes by Combined Nanoindentation and Mercury Intrusion Method

“…K-means is one of the most commonly used clustering analysis techniques. For example, Konstantopoulos [ 17 ] performed data labelling with unsupervised machine learning with k-means clustering to test novel Portland cement formulations with Carbon Nanotubes and intrinsic properties revelation. Due to the problems of the presupposition of mechanical properties of materials, complicated calculation, and uncertainty of initial value selection in the Gauss convolution method, Hou [ 18 ] applied cluster analysis to the study of nanoindentation of cement-based materials, and Krakowiak [ 19 ] compared the differences of different clustering methods and explained the applicability of clustering analysis method, which can increase the stability of the results.…”