Application of machine learning algorithm to assess the efficacy of varying industrial wastes and curing methods on strength development of geopolymer concrete

Abstract

The contribution of geopolymer concrete is irrefutable and imperative in the context of sustainable construction practices. The current study highlights the relevance of red mud and glass powder in conjunction with fly ash as the aluminosilicate source material for geopolymer concrete production. The performance of the designed binders was evaluated accentuating the function of fly ash replacement levels (10%-40%), curing time, and curing method (ambient, external exposure, and oven curing). Experimental results affirm that the mix containing 10% red mud and glass powder in equal proportions depicted the highest compressive strength at 28 days under external exposure (48 MPa) and oven curing (75 MPa). The study also revealed allowable leachability of geopolymer composites as per relevant standards. Analysing the 28 days compressive strength data using machine learning algorithms showed that the relationship between the dependent and the set of independent variables could be best predicted by a non-linear model, substantiated by the highest level of accuracy (98.79%).