Beneficiating ceramic waste in sustainable concrete pavement: multi-objective optimization using artificial neural networks

Le résumé fourni par la source

Mitigating natural resource depletion and ensuring that byproducts and wastes are reused effectively has become paramount for fostering sustainable development. Accordingly, this study explores using low-cost, readily available, and ecologically benign ceramic wastes to replace cement and fine/coarse aggregates in developing new types of mortar/concrete as pavement. The first phase of this work examines the durability, mechanical strength, and long-term performance of mortar/concrete that uses beneficiated ceramic waste as a secondary cementitious material and crushed ceramic waste as fine/coarse aggregates. To characterize the morphology, structure, and thermal properties of the proposed mixtures, scanning electron microscopy (SEM), X-ray diffraction (XRD), and Fourier-transform infrared spectroscopy (FTIR) were deployed. In the second phase, artificial neural networks (ANN) coupled with a Bayesian regularization training algorithm were tailored to estimate waste ceramic-based concrete’s compressive strength and CO 2 emissions. Sensitivity and regression analyses were developed to examine each parameter’s influence on the compressive strength and CO 2 emissions and facilitate the design of beneficiated ceramic-based concrete. Subsequently, a multi-objective optimization study was developed to identify the values of input predictor variables that maximize the predicted compressive strength and minimize CO 2 emissions. The findings show that using ceramic waste as a binder and source of fine/coarse aggregates vastly improved the mechanical strength performance of the mortar/concrete and reduced drying shrinkage and porosity. Moreover, given the favorable interactions between fine aggregates and ceramic waste, porosity decreased and microstructure enhanced.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.