Thermal conductivity of cementitious composites reinforced with graphene-based materials: An integrated approach combining machine learning with computational micromechanics
Arman Montazerian, Jan Arve Øverli, Stergios Goutianos
Abstract
Graphene-based materials (GMs) have significant potential for enhancing the thermal conductivity of cementitious composites. This study uses both machine learning (ML) and computational micromechanics models (CMMs) through a hybrid modelling approach to investigate the thermal conductivity of GM-reinforced cementitious composites (GRCCs). Accordingly, validated CMMs were used to train the ML models, which were then employed to predict the thermal conductivity of GRCCs. The results show that, among the parameters investigated, the volume fraction of GMs is the most influential factor in predicting the thermal conductivity of GRCCs, followed by their aspect ratio.
Topics & Concepts
MicromechanicsMaterials scienceThermal conductivityComposite materialVolume fractionCementitiousGrapheneConductivityThermalComposite numberCementNanotechnologyPhysical chemistryMeteorologyChemistryPhysicsInnovative concrete reinforcement materialsSmart Materials for ConstructionThermal properties of materials