Experimental and Artificial Neural Network-Based Study on the Sorptivity Characteristics of Geopolymer Concrete with Recycled Cementitious Materials and Basalt Fibres

Abstract

The environmental concerns regarding the production of the most widely consumed cement construction material have led to the need for developing sustainable alternatives. Using recycled industry waste products such as fly ash and slag via geopolymerisation has led to the development of geopolymer cement—an efficient replacement for ordinary Portland cement (OPC). Adopting geopolymer cement and concrete as a construction material reduces greenhouse gas and promotes the recycling of waste products. This study explores the suitability of a unique geopolymer concrete mix made of recycled cementitious materials including industry waste products such as fly ash, micro fly ash and slag for use in aggressive environments. Sorptivity tests are conducted to assess the durability of concrete and indicate the cementitious material’s ability to transmit water through the capillary forces. This study thus reports on the sorptivity characteristics of a newly developed self-compacting geopolymer concrete and two other fibre geopolymer concrete mixes containing 1% (by weight) of 12 mm- or 30 mm-long basalt fibres. The addition of basalt fibres indicated less water absorption and moisture ingress than the mix without fibres. The study used 18 specimens from three geopolymer concrete mixes, and the results showed that adding fibres improved the durability performance in terms of resistance to moisture ingress. Finally, an artificial neural network model is developed to predict the absorption rates of geopolymer concrete specimens using MATLAB. The prediction models reported excellent agreement between experimental and simulated datasets.

Artificial Intelligence Models for Predicting the Compressive Strength of Geopolymer Cements

“…The research findings obtained with the ANN model in this study are corroborated by previous studies, indicating that ANN demonstrates the predictive equation's adaptability to all types of modified geopolymer concrete mixtures in the future, thereby offering enhanced accuracy. The ANN model accurately predicts absorption rate as a function of binder content, basal fiber content, compressive strength, and changes in concrete mass [40]. Furthermore, other studies reveal that fly ash-based geopolymers utilizing a developed Neural Network model algorithm are potent tools for predicting geopolymer compressive strength [41].…”

Towards a Reliable Design of Geopolymer Concrete for Green Landscapes: A Comparative Study of Tree-Based and Regression-Based Models

“…Machine learning algorithms [10,31,32] Using machine learning algorithms, such as decision trees [33], random forests [34], neural networks [35], etc., the training of this model requires the input of relevant data of influencing factors and the characteristic data of known materials. Through the input of a large number of data sets, the model can predict the compressive strength of geopolymer concrete [36,37].…”

“…At the same time, machine learning models can provide more comprehensive predictions since the prediction of geopolymer compressive strength is a very complicated task. There are many influencing factors, so it is necessary to choose a suitable model or method for estimation and prediction [37]. Many factors can affect the properties of geopolymer concrete, and the relationship between these factors is often nonlinear [39].…”