Slump flow modeling of self-compacting conrete using smooth support vector regression (SSVR)

A new method of prediction based on smooth support vector regression (SSVR) is introduced to resolve the slump flow modelling of self-compacting concrete (SCC). The slump flow is a function of the content of all concrete ingredients, including cement, silica fume, water, superplasticizer, coarse and fine aggregate. In this paper, the basic ideas underlying SSVR are reviewed, and the potential of the SSVR for multiple regression (modelling) problems is demonstrated by applying the method to model of slump flow from experimental data. The results of experimentation indicate that SSVR has excellent performance on slump flow prediction. Compared with traditional prediction method such as second order regression, SSVR has much more accurate and effective to prediction of slump flow and it is very promising result.