Predictive Analysis of Compressive Strength of Geopolymer Concrete Using Deep Learning Techniques

Abstract

The prediction of compressive strength in geopolymer concrete is a critical aspect of optimizing construction materials for sustainable and durable infrastructure. In present study artificial neural network (ANN) is employed to forecast the compressive strength of geopolymer concrete (GPC). GPC, an ecofriendly alternative to traditional ordinary Portland cement (OPC) based concrete, poses challenges in predicting its compressive strength due to its complex composition. ANN is employed as a powerful tool to model the intricate relationships between various input parameters such as mix design, alkaline solution concentration and ratio, curing conditions, and the resulting compressive strength. The study involves collecting a comprehensive data set of geopolymer concrete mixtures and their corresponding compressive strengths, followed by training the ANN model. The model’s performance is evaluated and finetuned to ensure accurate predictions. This research aims to enhance our understanding of GPC behavior and provide a practical tool for engineers and construction professionals to optimize mix designs, and enhance sustainability by mixing over-engineering. The utilization of ANN demonstrates its potential as a valuable predictive tool for optimizing GPC formulations, ultimately contributing to more sustainable and resilient infrastructure.