Effect of mixture temperature on slump flow prediction of conventional concretes using artificial neural networks

Abstract

Hot weather concreting is unavoidable in the construction industry. Producing desirable workability in different temperatures needs careful recognition of temperature effects on workability. Several studies have shown that slump flow model of concrete is not only specified by the content of its ingredients, but that is also determined by fresh concrete placement temperature, ambient temperature, humidity, wind speed, etc. In this study slump flow prediction of conventional concretes using artificial neural network (ANN) models are presented. This paper discusses that to what extent inclusion of fresh concrete temperature can influence slump flow prediction of ANN models. It is also detailed that how variations in the “as placed” concrete temperature change the workability and influence of ingredients on workability. The results show that: (i) the ANN models of slump flow that incorporate the concrete temperature are a little more accurate and slightly has higher prediction precision than those they do not; and (ii) the ANN models are more precise in predicting slump flow rather than non-linear and linear regression models.