PSI - Issue 64

Veronica Bertolli et al. / Procedia Structural Integrity 64 (2024) 1033–1040 1037 Veronica Bertolli, Lesley H. Sneed, Francesco Focacci, Tommaso D’Antino/ Structural Integrity Procedia 00 (2019) 000–000 5

Fig. 2 Distribution of number of layers and textile equivalent thickness of the FRCM used in the database with respect to concrete compressive strength and fiber type.

Italian CNR-DT 215 (2018) and American ACI 549.4R (2020) guidelines were used to compute the analytical FRCM shear strength contribution, V f,an . Values of E FRCM and  fu were collected from clevis-grip tensile test results available in the literature, while values of  lim,conv were gathered from direct shear bond tests performed on the same FRCM available in the literature. If multiple layers of textile were used, the corresponding value of  lim,conv was either taken from bond tests with n textile layers or obtained with a linear interpolation of results associated with one and n layers. Further details can be found in Bertolli et al. (2024). The assessment of the models was performed using the following statistics:

f,exp,i V V

s N i



, , f an i

(9)

r



N

s



 2

r r

s N i N   i

1

(10)

CoV



r

s

where f,exp,i f an i r V V  is the ratio between the experimental and corresponding analytical prediction of V f , and N s is the total number of beams in the database. r represents the r i mean value, and the CoV provides indications of the model precision with respect to the mean value r . Values of r and CoV computed for specimens with the same fiber type and considering all specimens in the database are reported in Table 1. Considering all specimens in the database, values of r and CoV for CNR-DT 215 (2018) were 0.968 and 0.516, respectively, while values of r and CoV for ACI 549.4R (2020) were 2.883 and 0.941, respectively. Comparison between the two models considered in this work highlighted higher accuracy for the model proposed by the Italian guidelines (CNR-DT 215 (2018)). , , / i