Modeling of Steel-Reinforced Grout Composite System-to-Concrete Bond Capacity Using Artificial Neural Networks

Data Availability Statement

Acknowledgments

Notation

A_f

textile area;

b

bias [Eq. (2)];

b_c

mortar width;

b_f

textile width;

c_i

basis function [Eq. (4)];

E_f

elastic modulus of the dry fibers;

e1, e2

eccentricity values;

f_c

concrete compressive strength;

f_c,mat

mortar compressive strength;

f_ct

concrete tensile strength;

f_t,mat

mortar splitting strength;

L_eff

effective bonded length;

L_f

bonded length;

n

number of averaged data;

n_s

number of similar tested samples;

P*

peak load;

$P_{\mathrm{avg}}^{*}$

average peak load;

P_pl

debonding load (average value of the applied load along the plateau);

P_plavg

average debonding load;

P_t

maximum carrying load of the composite that corresponds to the tensile strength of the SRG;

P_u

load that corresponds to the global slip (s_fu);

P₁

load that corresponds to the global slip (s_f1);

R²

correlation index;

s_f

global slip;

s_fu

global slip that preceded the failure of the specimen;

s_f1

global slip at the first substantial drop in the load response after the peak load was reached;

t_f

equivalent thickness of the fiber sheet;

V_f

displacement rate.

w_i

weight [Eq. (2)];

x_i

numerical inputs [Eq. (2)];

Y_e

experimental value [Eqs. (7) and (8)];

Y_t

theoretical value [Eqs. (7) and (8)];

y

activation function [Eq. (3)];

y_i

neuron output in the hidden layer [Eq. (4)];

α

identity function [Eq. (5)];

γ

mass density of SRG;

μ

spread of the Gaussian function [Eq. (4)];

σ

strength associated with the bond capacity [Eq. (1)];

σ_lim,conv

conventional stress limit;

σ_uf

tensile strength of the fiber sheet; and

References