Prediction of compressive stress–strain behavior of hybrid steel–polyvinyl-alcohol fiber reinforced concrete response by fuzzy-logic approach

Abstract

The knowledge-based prediction of the compressive stress–strain behavior of concrete is of great importance for its optimization in structural applications. This paper proposes a fuzzy-logic numerical model to generate the complete stress–strain behavior of hybrid fiber reinforced concrete (HFRC). The model’s data population was comprised of 27 various HFRC mixtures. The training dataset for this mode included 18 mixtures tested by the authors, whereas the remaining checking dataset was collected from the literature. Comparing the proposed fuzzy-logic model with available prediction equations established the reasonability and strength of the proposed numerical model in predicting the complete stress–strain behavior of HFRC. The methodical inaccuracy analysis of the proposed model for calculating the elasticity modulus and strain at peak stress of HFRC validated the proposed procedure's prediction advantage compared to other analytical models. However, the suggested knowledge-based model showed the trend of insignificantly underestimating the elasticity modulus of the HFRC. The proposed model is likely limited to HRFC with strength in the 60–90 MPa range and containing a mixture of metallic and nonmetallic fibers.