Statistical interpretation of strength data for High-volume fly ash concrete – a new mix proportioning technique

Abstract

Posing its importance in building infrastructure, compressive strength is regarded as one of the most critical parameters to choose the grade of concrete. High performance concrete production not only involves huge physical effort and quality control, but also involves the incorporation of supplementary cementitious material in conjunction with high quality additives. In this present study, high volume fly ash content is used for high performance concrete mix. For any mix proportioning technique adopted, regression analysis is of utmost importance to validate the model. This current work covers detailed literature study on various forms of regression models developed for fly ash based concrete mixes. A new mix proportioning technique is adopted with cement replacement with class F fly ash as well partial addition of sand with fly ash as fine aggregate. In this present study, a linear regression analysis is conducted which produced excellent correlation coefficient for prediction of compressive strength at 28-days curing for high-volume fly ash concrete mixes. For the pulverized fuel ash substitution from 40%, 50%, 60% and 70% (cem:F_ash 1:1), improved workability properties were obtained with increased chemical admixture content (~1-2%). Good correlation coefficient (R²) values of 0.9831 or 98.3% are obtained at higher w/b ratios for 28-days curing. Trendline indicates the linear relationship with R² value of 0.9831 for higher w/b ratios and 0.974 for lower w/b ratios, suggesting a strong negative correlation. This signifies, a better model fit. Higher R² values of 0.98 signifies, good model fit, while R² values of 0.843 shows 15.7% variance due other properties in concrete such as reduced pozzolanic activity and fly ash reactivity at higher fly ash content beyond 40% and low water-binder ratios. Lower standard deviation values show that the data set is closely tight.