Discrete meso-scale modeling and simulation of shear response of scaled glass FRP reinforced concrete beams without stirrups

Evidence from load tests on concrete beams internally reinforced with corrosion-resistant glass fiber-reinforced polymer (GFRP) bars, and without shear reinforcement, shows that the sectional shear stress at failure drastically decreases at increasing effective depths. An important implication is that the strength and failure mode of typical laboratory-scaled specimens can be misleading if such results are used to extrapolate the response of larger members. In addition, the most accurate nominal shear strength algorithms are based on fundamentally different hypotheses on the governing shear-resisting mechanisms, namely aggregate interlocking and shear-compression fracture. Advanced computational modeling and simulation tools can aid with understanding the underlying mechanics of size-dependent shear response. This paper demonstrates the practical selection of the parameters of a Lattice Discrete Particle Model (LDPM) for a concrete for which only uniaxial compression test results are available, which is a typical case for specimens in existing databases. The LDPM was enlisted as it represents the physical heterogeneity of concrete, and incorporates constitutive laws (e.g., tension-softening and shear-compression fracture) that are critical to realistically simulate meso-scale friction and fracture damage mechanisms. The calibrated LDPM was used to build numerical models of slender GFRP RC beams with effective depth of 146 and 292 mm, for which evidence from actual load tests highlighted a size effect in the range 48–62%. The computational simulations yielded accurate estimates of strength, failure mode and load-deflection response irrespective of beam size. The output of the numerical simulations using the LDPM-based models was used to gain new insight into the contribution of different shear-resisting mechanisms to strength and size effect.