An Application of Artificial Neural Network to Predict the Compressive Strength of Concrete using Fly Ash and Stone Powder Waste Products in Central Vietnam

Abstract

In Central Vietnam, the traditional materials for making concrete are usually of natural origin. The overexploitation of these materials causes many adverse effects on the natural environment. Local industrial plants and quarries generate millions of tons of waste products such as fly ash and stone powder. However, when used for the partial replacement of cement and sand, these waste products can affect the compressive strength of concrete. Therefore, it is necessary to build models to predict compressive strength for this type of concrete. The paper aimed to apply artificial neural network models to predict the compressive strength of concrete using fly ash and stone powder waste products. The input of the ANN model includes six parameters: ultrasonic pulse velocity, wave amplitude attenuation ratio, and 4 parameters of concrete materials. Experimental data were obtained from 72 cubic specimens of different mixtures using available materials in Central Vietnam. These models allow predicting the 28-day compressive strength of concrete within the range of 9-62MPa (90-620daN/cm2). Furthermore, these models can predict compressive strength with any mixture. It is significant when re-evaluating whether the actual compressive strength value is as reliable as the one provided by the manufacturer.