Maiara G. Montaute, Hugo Luiz Oliveira, Thomaz E. T. Buttignol

Abstract: Topology Optimization (TO) is a mathematical method that searches for the ideal spatial distribution of materials based on the design domain, objective function, constraints and boundary conditions. There are different methods and softwares at hand, which can give distinct responses. Thus, the choice of the most appropriate algorithm and program is essential to provide a feasible solution. The aim of this work is to compare the results of the TO of simply supported rectangular Ultra-High-Performance Concrete (UHPFRC) beams, subjected to different static loads and spans ranging from 3 to 10m. The analyses include optimizing 3D parametric beams using Grasshopper's Algorithm-Aided Design (AAD), which is integrated with the Topos plugin, based on the SIMP model, and associated with Galapagos, a genetic algorithm plugin. This approach allows the automation search for the minimum volume of the rectangular beams. The second software used for TO processing is TOSCA, which is integrated with Abaqus/CAE, allowing optimization using the MIMP, RAMP and SIMP methodologies. Finally, the beams were processed in the Fusion software, from Autodesk Nastran In-CAD, which adopts the SIMP model. All the algorithms were used to provide the optimized geometry of the beams automatically to achieve the defined objectives of minimizing the volume and deformations. After the definition of the most suitable geometry, a non-linear analysis in Abaqus software is performed to check the maximum load bearing capacity (ULS) and the deflection (SLS). The results demonstrate a change in the failure mode, from flexural to a truss-like. Moreover, the optimized beams were able to maintain the initial load bearing capacity, without losing performance.

Keywords: Topology Optimization, Precast Beams, Uhpfrc, Fusion, Tosca, Simp.

Date Published: July 15, 2024
DOI: 10.11159/ijci.2024.010

View Article