Modularized Segmental Construction with CFRP-Reinforced Concrete Segments post-tensioned with CFRP-Tendons and Inline Quality Control by Computed Tomography

点击翻译按钮获取中文摘要

In the 1st phase of the project, the design and load-bearing behavior of modular segmental bridges with CFRP reinforcement and CFRP tendons as well as quality control using industrial computed tomography as a robust, quality-assured manufacturing method were investigated. Thus, the project contributed to stationary fabrication and flow production methods with CFRP-reinforced and -prestressed concrete elements, joining methods, the generation of digital 3D models as well as quality control for adaptive modularized construction in the SPP. In particular, the load-bearing capacity of profiled dry joints, their joint contact properties and the CT-based detectability of defects in the concrete were investigated. In the 2nd phase of the project, models for the load-bearing behavior of modular segmental bridges are to be developed on the basis of these results, as well as an overall design concept based on integrated quality assurance for precision prefabrication of the modules for large or medium series with robot-assisted generation of digital 3D models. These are to be validated on the basis of a bridge demonstrator consisting of modules suitable for flow production (weight max. 1t) and which is used both for analysis by means of computed tomography and for load-bearing capacity tests to determine the realistic, non-linear load-bearing behavior, taking into account the influences of segmental joints and CFRP tendons. Since the modules of the bridge demonstrator are larger and heavier than components that can be scanned by classical industrial CTs used in mechanical engineering, robotic CT will be used to generate 3D data of the stationary module by having the source and detector scan the edges and interfaces of the modules. By fusing X-ray-based 3D CT data with photogrammetric 3D surface data, digital 3D models are generated that simultaneously provide quality information about the surface condition (color, texture, homogeneity) and the internal structure of the component. Due to the much more complex trajectory of the X-ray source and detector in robot-based CT compared to laboratory-based CT scanners, the resulting 3D data contain significant artifacts, which will be corrected using deep neural networks (NN, Deep Learning). Considering the possible trajectories of robotic CT and the mutual relationships and requirements between robotic CT and structural design (cross-sectional geometry for good transmission ability), suitable recommendations for CT-based structural design shall be developed. Ultimately, a design concept based on the new possibilities of robotic CT for quality assurance as well as on FE models for the prediction of the load-bearing behavior for a material-optimized modularized segmental structure will be developed, which can be applied to segmental bridges as well as to other modularized building structures.