Comparative study of advanced computational techniques for estimating the compressive strength of UHPC

Abstract

The effect of raw materials on the compressive strength of concrete is a complex process, especially in the case of ultra-high-performance concrete (UHPC), where a higher number of inter-dependent parameters are involved in the strength development. In this era of digitalization, advanced machine learning methods are used to predict the material's mechanical characteristics because of their superior performance compared to conventional and nonlinear statistical regression models. Thus, the goal of the current study is to estimate the compressive strength of UHPC from the designed raw materials using advanced machine learning techniques. The compressive strength of UHPC is predicted from the 14 input parameters, i.e., cement, fly ash, slag, silica fume, nano-silica, limestone powder, sand, coarse aggregate, quartz powder, water, superplasticizer, PE fiber, steel fiber, and curing time. A total of eight machine learning models were compared that include multi-layer perceptron neural network (MLPNN), MLPNN Bootstrap aggregating (MLPNN-BA), MLPNN adaptive boosting (MLPNN-AB), Gradient boosting (GB), Decision tree (DT), DT Bootstrap aggregating (DT-BA), DT adaptive boosting (DT-AB) and Random Forest (RNF). The validation and performance evaluation of the above models were checked by using K-fold cross-validation, mean absolute error (MAE), root mean square error (RSME), coefficient of determination (R2), relative root mean square error (RRMSE), performance index (PI), and Nash Sutcliffe efficiency (NSE). The optimal model was selected based on the results of all statistical checks. It was found the ensembled machine learning models especially decision tree-based models outperform the neural network-based models with higher accuracy and low error. Thus, the recommended machine learning model is random forest having superior prediction capacity followed by DT Bootstrap aggregating and DT adaptive boosting.