Estimation Of Strength Properties Of Self Compacting Concrete By Using Artificial Intelligence And Machine Learning Techniques

Replacing a portion of concrete with Class F fly ash aids sustainable development and reduces the greenhouse effect. Developing an accurate predictive model for the compressive strength of self-compacting concrete (SCC) using Class F fly ash is crucial. This study evaluates various machine learning models using a dataset of 327 samples. Models include regression trees, support vector regression, Gaussian process regression, and artificial neural networks (ANNs). The ensemble of ANNs exhibits the highest accuracy, with a mean absolute error of 4.37 MPa and a correlation coefficient of 0.96. Additionally, simpler models like multi-genetic programming and individual regression trees perform comparably well. Self-compacting concrete, Class F fly ash, compressive strength, machine learning, artificial neural networks, regression trees, and Gaussian process regression are key terms in this study.