Mechanical performance and numerical simulation of high-performance steel fiber reinforced concrete

Abstract

This paper presents the outcome of the investigation of high-performance steel fiber reinforced concrete (HP-SFRC) with water to binder ratio (w:b) of 0.30- 0.40. The binder included cement replacement of 10 and 15% by silica fume. Fiber volume fractions (Vf) of 0, 0.5, 1.0 and 1.5% with aspect ratios of 80 and 53 were used. The study aims to present the effect of the inclusion of micro-silica and steel fibers on the mechanical performance of HP-SFRC. Experimental results show that moderate increase in compressive strength and significant improvement in flexural strength of HP-SFRC at Vf = 1.5% (reinforcing index = 3.88) with 15% micro-silica replacement. Empirical expressions were developed for compressive and flexural strength of HP-SFRC as a function of steel fiber volume fraction. The power relation between flexural tensile and compressive strengths was developed with an integral absolute error of 6.39. Further, the validity of the proposed models was tested against published data. A machine learning framework was established based on an adaptive neuro-fuzzy inference system (ANFIS) to predict the compressive strength of HP-SFRC mixes with higher accuracy. Finally, a multiple linear regression (MLR) model is proposed for the strength of HP-SFRC mixes and tested against published data. The validity of the MLR model is checked with experimental results and it is shown that the model has good prediction capability.