ANN-Based Prediction of Compressive Strength in Glass Wool Reinforced Brick

Abstract

The construction industry continuously seeks innovative materials and methodologies to enhance structural integrity while minimizing environmental impact. This study investigates the predictive capabilities of Artificial Neural Networks (ANN) in estimating the compressive strength of clay brick. Employing a dataset derived from comprehensive experimental trials encompassing varying compositions and curing conditions, an ANN model was developed and trained to predict the compressive strength of glass wool reinforced composite bricks. The inputs to the ANN comprised key parameters including the proportions of glass wool content, load at failure, area of cross-section and burning temperature. The model was optimized through iterative training processes to attain robustness and accuracy in predicting compressive strength. Subsequently, validation was performed using separate test datasets to evaluate the model’s generalization capacity. The results demonstrate the efficacy of the ANN model in accurately forecasting the compressive strength of glass wool reinforced clay brick. The analysis reveals nuanced correlations between glass wool content, load at failure, area of cross-section and burning temperature, and the resultant strength, shedding light on the intricate dynamics governing these composite materials. This ANN-based predictive approach presents a useful tool for engineers and stakeholders in the construction industry to anticipate and optimize the compressive strength of glass wool reinforced clay bricks. Furthermore, the findings contribute to advancing the understanding of these novel composite materials, fostering sustainable and resilient construction practices.