Electrical resistivity-Compressive strength predictions for normal strength concrete with waste steel slag as a coarse aggregate replacement using various analytical models

Concrete is a composite material that is highly used in construction fields. Steel slag (SS) is a molten liquid melt of silicates and oxides, is a by-product of the steel-making process, and solidifies upon cooling. It is a complex solution of Silicates and Oxides. From an environmental standpoint and to save the environment and natural resources, steel slag recovery conserves natural resources and frees up space in landfills. Steel slag as waste materials has been used in concrete as a partial replacement with fine (sand) and coarse aggregate (gravel). A total of 338 data points were collected, analyzed, and modeled. The most effective factors affecting the compressive strength (CS) of concrete incorporated with steel slag replacement were considered during the modeling process. The cement content was ranged from 237.35 to 550 kg/m(3), curing time 1-180 days, water/cement ratio ranged between 0.3 and 0.872, steel slag content varied between 0 - 1196 kg/m(3), fine aggregate content ranged between 175.5 - 1285 kg/m(3) and coarse aggregate content (natural aggregate) varied between 0 - 1253.75 kg/m(3). Furthermore, 58 data were collected to analyze and model the effect of steel slag on the electrical resistivity (ER) of normal concrete. An Artificial Neural Network (ANN), a Multi Logistic Regression model (MLR), a Full Quadratic model (FQ), and an M5P-tree model were employed in this study to forecast the compressive strength of normal strength concrete (CS ranged between 10 and 55 MPa) with steel slag aggregate replacement, and Full quadratic model was applied to predict the ER of normal concrete (ER varied between 23.98 and 1440 & OHM;.m). Finally, the correlation between compressive strength and electrical resistivity was investigated through different models. Based on data from the literature, the steel slag content was increased the compressive strength and lowered the ER of concrete. According to statistical tool assessments such as objective function (OBJ), scatter index. Taylar diagram, the ANN model with the lowest root mean square error (RMSE) performed better than the other models in predicting the compressive strength. FQ model as a reliable mathematical model can be applied to forecast the ER of normal concrete due to the high coefficient of determination R2 is 0.91.

Automated summary

Steel slag can be used as a replacement material in concrete to improve compressive strength and reduce electrical resistivity. Different models have been employed to predict concrete performance, and the most accurate prediction was achieved using an Artificial Neural Network (ANN) model for compressive strength and a Full Quadratic model for electrical resistivity.