Development of a Reconstruction Methodology Based on X-Ray Computed Tomography to Generate Realistic 3D Concrete Microstructures in MOSAIC

Light Water Reactor Sustainability: Program Materials Research Pathway (FY22 Technical Program Plan)

“…The approach uses a modified version of the fully convolutional network (FCN) U-Net to label images and identify aggregates, cement paste, and pores or background based on a provided network training data set. The performance evaluation of the network revealed a successful application of this approach [46].…”

“…Consequently, damage is mainly governed by structural constraints caused by the biaxial compression loading near the reactor cavity. In FY 2021, the first attempt to construct 3D microstructures using x-ray computed tomography (XCT) of concrete for input to MOSAIC simulations was developed [45]. The development and validation of MOSAIC-3D is critically important to overcome the limitations of MOSAIC-2D.…”

Light Water Reactor Sustainability Program Materials Research Pathway Technical Program Plan

“…Moreover, in FY 2021, the first attempt to construct 3D microstructures using x-ray computed tomography (XCT) of concrete and its components based on the successful development, application, and validation of the FFT-based MOSAIC using realistic 2D concrete microstructures [39]. The development and validation of MOSAIC-3D is critically important to overcome the limitations of MOSAIC-2D.…”