Compressive strength analysis of fly ash-based geopolymer concrete using machine learning approach

Fly ash has been widely used in studies to reduce the need for cement in concrete mixtures. This study used machine learning modeling using Python Jupyter to predict the compressive strength (CS) of concrete containing fly ash. An ANN model with seven parameters as input data (specimen age at the time of testing, cement, fine aggregate, coarse aggregate, superplasticizer, water, and fly ash) is developed to predict CS. A total of ۴۶۰ datasets are used for ANN modeling after an extensive review of relevant published articles. Machine learning performance was evaluated using a set of two metrics, Pearson correlation coefficient (R) and root mean square error (RMSE). The evaluation output shows that the predicted results correlate with the actual results of the experiments. The proposed models can be used to make a standard mixture and to design the mixture proportions of geopolymer concrete based on fly ash.

Keywords:

Keywords: fly ash content , Concrete mixtures , artificial neural networks , compressive strength prediction , Statistical analysis