Predictive modelling of mechanical properties of concrete using machine learning with secondary treated waste water and fly ash

Abstract

This study investigates the utilization of secondary treated wastewater and fly ash in concrete production, focusing on modelling mechanical properties using machine learning models. Sixteen concrete mixtures were prepared with tap water and three types of secondary treated wastewater, varying the fly ash proportions (0 %, 10 %, 20 %, and 30 %). Workability, compressive strength, split tensile strength, and flexural strength were assessed for each mixture. Five machine learning models Linear Regression, LASSO Regression, Decision Tree Regression, Random Forest Regression, and Multi-Layer Perceptron were used to predict concrete's mechanical properties. The results show that M30 grade concrete can be effectively produced using secondary treated wastewater and fly ash, presenting a promising strategy for more sustainable concrete production by reducing freshwater usage and incorporating fly ash as a supplementary cementitious material. Notably, the Random Forest Regressor demonstrated superior prediction accuracy for compressive strength, outperforming the other models and proving to be an invaluable tool for optimizing concrete mix designs. Its ability to reliably predict concrete strength properties ensures higher accuracy in mix design formulation, which is critical for achieving desired performance while minimizing material waste. From a sustainability perspective, using secondary treated wastewater in concrete production significantly reduces the demand for freshwater, conserving this precious resource. Incorporating fly ash, a byproduct of coal combustion, not only enhances concrete properties but also helps divert industrial waste from landfills, reducing environmental impact. The application of machine learning models, especially the Random Forest Regressor, allows for more precise and efficient mix designs, further contributing to the sustainability of concrete production. This approach offers substantial environmental benefits by reducing water usage, promoting recycling of industrial byproducts, and improving the overall efficiency of concrete manufacturing processes. • Utilization of secondary treated wastewater and fly ash in concrete production. • Modelling of mechanical properties using machine learning models. • M30 grade concrete can be effectively produced using secondary treated wastewater and fly ash. • Research highlights using treated wastewater, fly ash, and machine learning to predict concrete properties.

Topics & Concepts