Refined predictive models for compressive and flexural strengths of FRC: a comprehensive study on synthetic and hybrid fibers

Abstract

The current research on synthetic and hybrid fiber-reinforced concrete (FRC) is extensive; nevertheless, there is a need for robust models to assess their mechanical strength. This study comprehensively evaluates test results from 192 FRC specimens, focusing on compressive and flexural strengths. This investigation thoroughly analyzed a range of diverse concrete formulations, incorporating two types of steel fibers, polypropylene, and polyvinyl alcohol fibers, in various configurations (mono-fiber and hybrid systems). This approach aims to investigate the impact of different fiber combinations on the mechanical properties of FRC, providing insights into their synergistic effects. Additionally, the research studied several available models for predicting compressive and flexural strengths. Moreover, the study proposes a refined model employing a multiple-regression analysis approach. The findings suggest that the inclusion of hybrid steel fibrous systems notably improves compressive strength (3.4–8.2%). Hybrid steel-synthetic fibers in FRC also have positive effects (+ 2.2–4.6%), while a mono-synthetic fibrous system shows a potentially negative impact. Significant enhancements in flexural strength (up to 103.4%) were observed in hybrid steel fiber-based SFRC. However, certain mixtures in the synthetic-steel fiber series displayed insignificant strength gains, emphasizing the necessity for an optimized hybrid fibrous system. The study reveals the varying predictive capabilities of the studied available models for compressive strength, with clear limitations in accurately predicting flexural strength for synthetic-based FRC. The proposed flexural strength model exhibits significant concordance with test data, with predicted-tested value ratios within the range of 0.89–1.16. Moreover, these models exhibited high predictive accuracy for compressive strength across 53 concrete mixtures from various independent studies, achieving an average predicted-to-actual ratio of 1.0 and a notably low coefficient of variation (CV) of 14.7%. Conversely, the predictions for flexural strength were more variable, with an average ratio of 1.14 and a higher CV of 30.1%. The precision and reliability of the proposed flexural strength model underscore its efficacy for diverse fibrous systems.