The computational and simulation analysis of pullout fiber reinforced concrete was investigated. The finite element analysis was used to make this modeling and analysis on this reinforced system and three parts (concrete matrix, the placed fiber reinforcement polymers (FRP), and resin layer) were studied. A constant load was directly applied on the free end of placed FRP and the deformation, von Mises stress, displacement, and strain of these three analyzed parts were obtained. Meanwhile, the specimen system of bonding strength and strain was calculated by the method of ABAQUS. The results showed that, with the constant load, the von Mises stress, deformation, and strain appeared in these three parts, and the maximum values in both FRP and resin layer were shown at the free end side, which provides an accurate description of the rupture mode.
Concrete is the most widely used construction material in large quantities for its low cost and wide availability [
Prior works on FRP reinforced concrete have focused on the finite element analysis modeling and simulation in the literature [
In this paper, the finite element analysis [
In order to confirm the analysis of the simulation program and the further studying about the relationship between bonding and slip, a finite element analysis was performed by using ABAQUS 6.81 to calculate and simulate the pullout FRP/concrete system. In the pullout FRP/concrete model, the main component includes four parts: concrete, FRP, resin, and the bond interface between concrete and FRP. All components were modeled by using 8node linear brick, which reduced integration and hourglass control (C3D8R).
As a simplified model of pullout FRP/concrete system, which was shown in Figure
Young’s modulus and Poisson’s ratio of three analyzed parts.
Materials  Young’s modulus/GPa  Poisson’s ratio 

Concrete matrix 

0.3 
FRP 

0.2 
Resin layer 

0.39 
Simplified model of FRPconcrete.
Before the model procedures, we assume that the bending effect of FRP is ignored. Meanwhile, only the shear force appears in adhesive layer and the size of each part still stays in a constant value and no deformation occurred.
For each component of this modeled system, the stressstrain curve of concrete follows the mathematical model investigated by Todeschini et al. [
Define a 3D concrete block with a sized groove placed in the middle of the matrix. The size of groove is
Define a 3D deformable FRP plate
Define two layers of the deformable coated resin
Define interfacial bond between FRP plate and concrete and FRP plate and resin layer by tying constraint of two adjacent surfaces. In this modeling, the tie constraint surfaces include concretefirst resin layer, first resin layerFRP, FRPsecond resin layer, and second resin layerconcrete. The defined model is shown in Figure
Summary of testing results.
Specimen 






Failure 

30 MPa25010  250  10.29  1.22  30  31,520  26.6  D 
The detailed dimensions of each three parts.
Sample 











30 MPa25010  350  300  180  350  1.22  10.29  250  1  10.29  26600 
The detailed reinforced system model.
As one important step in this modeling, a detailed meshing [
Mesh of concrete matrix, FRP plate, resin layer, and the reinforced concrete system.
The loading and boundary condition of reinforced concrete system.
In this simulation, a directly nonlinear analysis technique was employed and this technique followed the method of NewtonRapson. In the modeling procedures, the system stayed in a static loading condition. Automatic time step was applied with set 1. The maximum number was 100 and the increment size included the initial value 1, the minimum value 1E05, and the maximum value 1.
In order to satisfy the accuracy of the model, the bond interface characteristics of the analytical model were calculated by using the method of finite element analysis. In the Seracino et al. [
As calculated in (
Calculation for each specimen test by Abaqus software.
Specimen  Bond strength (kN) 


30 MPa10010  20.4  0.009133 
30 MPa15010  23.2  0.010213 
30 MPa20010  27.9  0.012238 
30 MPa25010  26.6  0.0118 
30 MPa30010  26.0  0.011452 
30 MPa35010  23.0  0.010106 
With the constant loading, a deformation was obtained in FRP plate. No deformation appeared in the area of
Deformation of FRP in this modeling.
von Mises stress [
von Mises stress of concrete, FRP plate, and resin layer.
The displacement showed the degree of deformation of concrete matrix, FRP plate, and resin layer, which indirectly reflected the bonding strength. The maximum displacement of tested three parts (Figure
Displacement of concrete, FRP plate, and resin layer.
The simulated strain showed the deformation resistance ability, which indirectly reflected the displacement and bonding strength. The same distribution trends were obtained in Figure
Strain of concrete, FRP plate, and resin layer.
In this paper, the computational and modeling analysis of the pullout FRP/concrete system was studied systematically. A finite element analysis was used in this modeling procedure. Each specimen system of bonding strength and strain was calculated by the method of ABAQUS. All the three parts, concrete matrix, FRP, and resin layer, were studied in this analysis individually. A constant load (26600 N) was applied on the free side of the placed FRP. Deformation, von Mises stress, displacement, and strain of each individual part were obtained and the maximum values all occurred at the edge of the contact point. Meanwhile, the values decreased with the deeper groove of concrete, which showed that the higher bonding strength was gained in the deeper groove, and the contact point was the weakest zone in this pullout FRP/concrete system.
Further researches are needed to obtain a deeper analysis of pullout FRP reinforced concrete. Also, the detailed pullingout process (elastic stage, elastic softening stage, debonding stage, and softeningdebonding stage) and slip and shear stress at the interfacial bond shall be explored. We believe that our results at least in the trend are helpful for the research of FRP reinforced concrete system.
The authors declare that there is no conflict of interests regarding the publication of this paper.
The authors would like to express appreciation for the financial support by the Natural Science Foundation of China (51178361).