Prediction of the strength of mineral-addition concrete using regression analysis

Abstract

The aim of this study was to develop a method for estimating concrete strength through non-destructive testing (NDT) methods. 27 different concrete mixtures were prepared with additions of high-volume fly ash and blast-furnace slag in place of Portland cement. Tests were then conducted to determine compressive strength (CS), splitting tensile strength (STS), flexural tensile strength (FTS), rebound number (RN) and ultrasonic pulse velocity (UPV) at 3, 7, 28, 90 and 180 days after preparation. The well-known traditional RN and UPV tests were used in this study, and correlations between NDT data and the strength properties of concrete were statistically evaluated using simple regression analysis. The analysis revealed exponential, geometric and power relationships between RN and CS, STS and FTS, respectively, with moderate correlation coefficients of 0·73, 0·72 and 0·72. Exponential and hyperbolic relationships were found between UPV and CS, STS and FTS, yielding correlation coefficients of 0·85, 0·85 and 0·79, respectively. Taking advantage of the fact that UPV and RN are related to different concrete properties, a combined method was applied to estimate concrete strength. The correlation coefficients of the models for CS, STS and FTS were 0·90, 0·84 and 0·88, respectively. The accuracy of the model in predicting CS was verified by statistical tests, and it was found to be reliable in practical situations.