PSI - Issue 70

Rachit Sharma et al. / Procedia Structural Integrity 70 (2025) 386–393 Sharma and Laskar/ Structural Integrity Procedia 00 (2025) 000 – 000

387

2

1. Introduction Fiber reinforced polymer is gaining traction over steel reinforcement because of their non-corrosive properties in coastal environment. However, the use of FRP rebars as flexural reinforcement leads to modifications in certain shear transfer mechanisms Nakamura (1995); Razaqpur et al. (2004); Johnson and Sheikh (2016). It is because the shear capacity for FPR-RV beams is affected by several influencing parameters such as beam geometry, span-to-depth ratio, reinforcement resistance factor and concrete compressive strength Machial et al. (2016); Nasrollahzadeh and Aghamohammadi (2018). Unlike steel reinforcement, which undergoes yielding, shear design for FRP rebars requires a high degree of confidence, as their brittle failure can lead to severe consequences. As a result, the current design code equation tends to yield conservative estimations of shear strength of FRP-reinforced concrete (FRP-RC) members. This leads to higher reinforcement ratios, resulting in reinforcement congestion and an overall increase in construction costs. Tottori and Wakui (1993). Table 1 presents the shear calculation method for concrete beams reinforced with FRP bars. The ACI 440.1 R-15 (2015) and GB 50608 – 2020 (2020) do not account for the shear-span to depth ratio effect and also size effect. CSA (Canadian Standards Association) (2012) and JSCE (Japan Society of Civil Engineers) (1997) accounts for size effect through factor k q and respectively. However, ( a/d ) ratio is only considered in CSA-S806-12, indicating that other mentioned methods predict zero shear strength for flexural members with no longitudinal reinforcement.

Table 1. Shear calculation method for concrete beams reinforced with FRP bars.

= ⁄ = ⁄

Method

Concrete shear contribution

Parameters considered

ACI 440.1 R-15 (2015)

 

  

( ) f    + − 2 f f

( ) c f

1/2

0.4 2 

V

bd

=

c

GB 50608–2020 (2020)

 

  

( ) f    + − 2 f f

( ) c f

1/2

0.86 2 

V

bd

=

c

1/2  ; ) 1/3 f   

CSA (Canadian Standards Association) (2012)

( ) 1/3

0.05

V

m r s a c k k k k f 

bd

=

  

V

(

) 1/3

c

v

1 E  = +

k

f

k

d

=

r

f

f

m

M

f k E  = + when d ≤ 300 mm 1.0 s k = , d ≥ 300 mm 2.5 f a f V k d M    =        , when a/d > 2.5 1.0 a k = ( 1 s f

JSCE (Japan Society of Civil Engineers) (1997)

/ f bd     = d p n vcd b

V

1/3 0.2( ) 0.72 f 

f

MPa

=

c

vcd

c

4 1000/  1/3 (100 / ) 1.5 f s E E   1; n for no axial force  = ; 1.5 d  = d p f  =

The design methods exhibit diverse approaches and lack a unified agreement across different codes regarding the parameters influencing shear strength. The methods continue to evolve and remain a subject of ongoing research and exploration. Therefore, establishing a reliable shear prediction model for FRP-RC sections is crucial. The data-driven machine learning (ML) models can offer a robust framework for capturing the intricate relationships between response variables and their predictors. Recently, ML techniques have successfully addressed several complex structural challenges Alam and Asce (2023); Kuo et al. (2024). A recent study by Alam et al. (2021) employed a hybrid Bayesian optimization algorithm and support vector regression (BOA-SVR) model to achieve reliable shear strength predictions for FRP-RC simply supported beams and one-way slabs, yielding a correlation coefficient (R) of 0.977 and a fractional : modular ratio; : elastic moduli of FRP; : elastic moduli of concrete; : longitudinal reinforcement area; : section width; : section effective depth; : longitudinal reinforcement ratio; V f : ultimate shear; M f : ultimate moment 1.3 b  =