Railway Infrastructure Asset Analysis Using A Global-Local Approach

Abstract

The global-local modeling methodology is used to evaluate specific regions of a structural system. A global model represents the entire system under study, while critical domains are assessed in detail by a submodeling routine employing more refined models. The advantages of this hierarchical modeling are related to reducing the need for complex transition regions in solid elements and the versatility in testing different geometries in the submodel region. Furthermore, it allows for a reduction in the computational power required to solve the problem. However, ensuring a good transfer of the boundary conditions between the different models is essential. For large and complex infrastructures, such as bridges, numerical analyses sometimes become time-consuming when more complex evaluations are required. Therefore, using simplified models to reproduce global behavior and more complex modeling strategies in critical locations can be an alternative to comply with this requirement. The present paper aims to employ a global-local approach to analyze a reinforced concrete railway bridge. Thus, the global and local models employed numerical analyses using solid tetrahedral finite elements. The local region presented a greater mesh discretization for the submodeling. Additionally, the local model allows inserting steel reinforcement details and specific constitutive laws for the materials utilized in the described region of the railway infrastructure. The obtained results enable the evaluation of the formation and propagation of cracks and the identification of damages located in the structural elements with greater precision. The methodology can improve condition assessment and support the inspection and maintenance of critical infrastructure assets.