ABOUT THE INTERNATIONAL CONCRETE ABSTRACTS PORTAL

  • The International Concrete Abstracts Portal is an ACI led collaboration with leading technical organizations from within the international concrete industry and offers the most comprehensive collection of published concrete abstracts.

International Concrete Abstracts Portal

  


Title: Shear Strength Equation and Database for High-Strength High-Performance Fiber-Reinforced Concrete and Ultra- High-Performance Concrete Beams without Stirrups (Open Source)

Author(s): Manuel Bermudez and Chung-Chan Hung

Publication: Structural Journal

Volume: 121

Issue: 4

Appears on pages(s): 185-195

Keywords: beam shape; closed-form equation; fiber distribution; highperformance fiber-reinforced concrete (HPFRC); hybrid fibers; machine learning (ML); shear-transfer mechanism; size effect; ultra-highperformance concrete (UHPC)

DOI: 10.14359/51740716

Date: 7/1/2024

Abstract:
The study presented a shear strength equation for high-strength high-performance fiber-reinforced concrete (HS-HPFRC), including ultra-high-performance concrete (UHPC). This equation was designed for straightforward implementation, catering to the regular tasks of engineers. It considers various influences on shear transfer mechanisms, including fiber bridging, fiber distribution, dowel action, cross-sectional shapes, and beam size effects. The equation does not rely on uniaxial tensile tests or inverse analysis of flexural tests; instead, it considers the statistical impact of fibers on shear strength. To generate the coefficients for this semi-empirical closed-form equation, an evaluation database of 118 HS-HPFRC and UHPC beams was constructed. The evaluation results revealed that the proposed equation has a mean of 1.00 and a correlation coefficient of 0.92, indicating low variation and high predictive accuracy. Furthermore, it outperformed existing equations and matched the accuracy of the machine learning (ML)-based models including support vector machines (SVM), random forest (RF), and artificial neural network (ANN), despite its comparatively simpler expression.