Numerisk Analys av Spjälkning i Betong

Sammanfattning

Splitting refers to cracking in concrete caused by transverse tensile stresses and is typically mitigated through reinforcement of the concrete. In Boverkets handbok om betongkonstruktioner, BBK04 threshold values are provided for when splitting is expected to occur and when splitting reinforcement is required. In 2011, Swedish design practice transitioned to Eurokod 2, in which guidance and design requirements regarding the phenomenon of splitting are not explicitly provided. Furthermore, experience regarding the numerical modelling of splitting remains limited. The aim of this study is to evaluate the threshold values presented in BBK04 regarding when splitting cracks are expected to occur, as well as the influence of the reinforcement prescribed by the strut-and-tie models in Eurocode on the cracking behaviour and load-bearing capacity of concrete. To enable this analysis, modelling choices affecting the propagation of splitting cracks are evaluated. In addition, a method for modelling the bond-slip between concrete and reinforcement using springs is developed and assessed. The analysis comprised three finite element models assembled in the software Brigade/Plus using a parametric Python script. Initially, a reinforced deep beam reinforced was evaluated with respect to flexural cracking. The results were compared with an established strut-and-tie model in order to validate the modelling approach. The remaining two models were analysed with respect to splitting cracks and the influence of reinforcement designed according to Eurocode strut-and-tie models. The reinforcement was modelled both wit and without bond-slip behaviour. The analyses demonstrated that modelling splitting failure is associated with significant complexity. The modelling choices found to have the greatest influence on the splitting crack propagation were the width of the loaded area, the choice of boundary conditions and the bond-slip behaviour. The analyses indicated that the BBK04 guidelines for when splitting is expected to occur are conservative. The ductility of the studied reinforced concrete structure increased when splitting reinforcement was introduced. The splitting reinforcement also influenced the crack propagation. Regarding the modelling approach for the bond between concrete and reinforcement, the results indicate that more realistic behaviour is achieved by connecting the reinforcement to the concrete using springs, as compared to an approach where the reinforcement is fully constrained to the concrete. The stiffness and anchorage of the springs had a significant influence on the results.