AUTHOR=Khan Mohsin Ali , Zafar Adeel , Farooq Furqan , Javed Muhammad Faisal , Alyousef Rayed , Alabduljabbar Hisham , Khan M. Ijaz 

TITLE=Geopolymer Concrete Compressive Strength via Artificial Neural Network, Adaptive Neuro Fuzzy Interface System, and Gene Expression Programming With K-Fold Cross Validation

JOURNAL=Frontiers in Materials

VOLUME=Volume 8 - 2021

YEAR=2021

URL=https://www.frontiersin.org/journals/materials/articles/10.3389/fmats.2021.621163

DOI=10.3389/fmats.2021.621163

ISSN=2296-8016

ABSTRACT=The ultrafine fly-ash (FA) is a hazardous material collected from coal productions, which has been proficiently employed for the manufacturing of geo-polymer concrete (GPC). In this paper, the three artificial intelligence (AI) techniques namely; artificial neural network (ANN), adaptive neuro-fuzzy interface (ANFIS), and gene expression programming (GEP) are used to establish a reliable and accurate model to estimate the compressive strength (f_c^') of fly-ash based geopolymer concrete (FGPC). A database of 298 instances is developed from the peer-reviewed published work. The database consists of the ten most prominent explanatory variables and f_c^' of FGPC as a response parameter. The statistical error checks and criteria suggested in the literature are considered for the verification of the predictive strength of the models. The statistical measures considered in this study are MAE, RSE, RMSE, RRMSE, R, and performance index (ρ). These checks verify that ANFIS predictive model gives an outstanding performance followed by GEP and ANN predictive models. In the validation stage, the coefficient of correlation (R) for ANFIS, GEP, and ANN model is 0.9783, 0.9643, and 0.9314 respectively. All three models also fulfill the external verification criterion suggested in the literature. Generally, GEP predictive model is ideal as it delivers a simplistic and easy mathematical equation for future use. The k-fold cross-validation (CV) of the GEP model is also conducted, which verifies the robustness of the GEP predictive model. Furthermore, the parametric study is carried via proposed GEP expression. This confirms that the GEP model accurately covers the influence of all the explanatory variables used for the prediction of f_c^' of FGPC. Thus, the proposed GEP equation can be used in the preliminary design of FGPC.