5.2 Comparison between simulation results and laboratory test results with microstructure incorporated for the concrete target

As previously mentioned, the 6″ <sup>12</sup>″ concrete cylinders were X-ray CT scanned and cross-section images showing their internal structure were obtained for each specimen. An image analysis code has been developed to reconstruct the internal structure of the target specimen using the x-ray CT slices. Thus, different constituents of concrete are identified respectively, based on the gray levels of a cross-section image. In addition to the reconstruction of the internal structure, the program also maps each pixel of the image onto the mesh of the digital model as shown in Figure 13. For each of the two 6″ <sup>12</sup>″concrete specimens, 100 slices

Concrete Microstructure Characterization and Performance DOI: http://dx.doi.org/10.5772/intechopen.90500

Figure 13. Internal structure reconstruction of the concrete specimen.

5.1 Comparison between FEM simulation results and laboratory test results

5.2 Comparison between simulation results and laboratory test results with

Test # Target type Test penetration depth (mm) Simulated penetration depth (mm)

1 Cyl (11″ <sup>11</sup>″) <sup>80</sup> <sup>84</sup> 2 Cyl (11″ <sup>11</sup>″) <sup>96</sup> <sup>83</sup> 6 Cyl (11″ <sup>11</sup>″) <sup>87</sup> <sup>150</sup> 7 Cyl (6″ <sup>12</sup>″) <sup>27</sup> <sup>18</sup> 8 Cyl (6″ <sup>12</sup>″) <sup>51</sup> <sup>106</sup>

As previously mentioned, the 6″ <sup>12</sup>″ concrete cylinders were X-ray CT scanned and cross-section images showing their internal structure were obtained for each specimen. An image analysis code has been developed to reconstruct the internal structure of the target specimen using the x-ray CT slices. Thus, different constituents of concrete are identified respectively, based on the gray levels of a cross-section image. In addition to the reconstruction of the internal structure, the program also maps each pixel of the image onto the mesh of the digital model as shown in Figure 13. For each of the two 6″ <sup>12</sup>″concrete specimens, 100 slices

microstructure incorporated for the concrete target

Penetration depth comparison between test data and simulation (Zhou et al., 2009).

It is worth mentioning that both the mass and diameter of the projectiles influence the penetration depths. However, the major factor to determine the penetration depth is the velocity of the projectile. For example, in tests 7 and 8, the two projectiles have similar masses and diameters, but the projectile with higher velocity (405 m/s) has a penetration depth of 51 mm which is almost twice the depth of the one with lower velocity (360.5 m/s). This fact is also revealed by the simulation results, pertaining to the same tests, in which the projectile having a higher velocity has a 106 mm penetration depth, while the one with lower velocity has a 51 mm penetration depth. The major reason for the inconsistency between the simulation results and the test results is that the material properties including microstructure

without considering the microstructure of the target

8000 psi cylinder impacted by 30 mm diameter and 218 g projectile at v = 360 m/s.

Figure 12.

Compressive Strength of Concrete

Table 3.

42

were not varied for different materials (Table 3).

(cross-section images) were stacked together and processed to generate their respective digital specimen. Elements pertaining to different components of the mixture (i.e., aggregates and cement paste) were assigned different material properties in the simulation. The aggregates were treated as elastic material with high elastic stiffness, whereas the cement paste was treated as an elasto-plastic material with low elastic stiffness and shear damage factor to control the damage of the material.
