Prediction of compressive strength of nano-silica concrete by using random forest algorithm

The prediction of compressive strength in concrete is an important part of civil engineering because it influences structural design and durability. Recent improvements have brought nano-silica as an addition, improving concrete’s mechanical qualities. This study investigates the use of the Random Forest (RF) method to estimate the compressive strength of nano-silica concrete, taking advantage of its ability to manage complicated, non-linear interactions among factors. We used a large dataset with varying concentrations of nano-silica and other concrete mix components. Normalisation and feature selection were used as preprocessing techniques to improve the model’s performance. The RF model, which is known for its resilience and accuracy, was trained and verified, displaying better predictive performance than typical linear regression models. The results showed that the RF method could efficiently capture the subtle correlations between nano-silica content and compressive strength, resulting in accurate predictions over a wide range of concrete compositions. This prediction model is a vital tool for civil engineers, allowing for more precise and efficient design of nano-silica concrete mixtures, resulting in stronger and more durable projects.

A complete dataset with varying concentrations of nano-silica and other concrete mix components was used. Preprocessing techniques, such as normalisation and feature selection, were used to improve the model's predicted accuracy. The RF model was trained and tested, demonstrating higher performance over established prediction methods like linear regression. The results showed that the RF algorithm could successfully forecast the compressive strength of nano-silica concrete while also capturing the complex interactions between the mix ingredients. This predictive capability is critical for civil engineers since it provides a powerful tool for creating concrete mixtures with improved performance characteristics. To summarise, incorporating machine learning approaches such as the Random Forest algorithm represents a substantial development in forecasting the compressive strength of nano-silica concrete, fostering innovation and dependability in the field of construction materials.