Barthold, Franz-JosephMolod, Mohammad Amin Esmail2021-04-162021-04-162021http://hdl.handle.net/2003/40159http://dx.doi.org/10.17877/DE290R-22031Column-beam joints are one of the most critical zones of concrete structures, especially under unpredicted heavy loads and lateral loads such as seismic. Failure of the joints can even lead to failure of structures in their entirety. The low strength capacity of concrete is a reason of sensitivity of the region. Shape memory alloy (SMA) plates can be employed in order to overcome this weakness and increase the stiffness of joints in existing structures. SMA is a smart material whose functionality, workability and its self-healing feature are under investigation by scientists in the field of structural engineering. In fact, there are two types of alloy: i) superelastic shape memory alloy and ii) shape memory effect that is sensitive to temperature, but it is out of the topic of the research. However, a superelastic form is the most common type of alloy in the field of structural engineering that can be used not only as external reinforcement bolted to the concrete surface but also as internal reinforcement embedded within the concrete elements. The author of this numerical research attempted to implement a plate form of the alloy as external reinforcement to increase stiffness and ductility of the joint. To do so, an experimentally investigated concrete column-beam joint has been modelled in Ansys, and it was loaded under a large number of randomly selected load combinations. The plate initially was designed with a uniform thickness and length in the plastic hinge region of the joint under the critical load combination. Then, probabilistic analysis was carried out to optimize the plate’s thickness. To that end, the stress values of thirty-five predefined nodes on the plate surface were recorded under each load combinations. Results were imported into MATLAB software to run the probabilistic analysis and specifying 0.95 quantile of the stored stresses of the nodes. Design optimization was also carried out based on the probabilistic results in order to design the thickness of the plate at different control nodes. During the course of the research, a set of necessary additional trials have been carried out, as for example with regards to the proper Ansys element type selection for reinforced concrete, determination of limit state functions, and to assess the most suitable parallel processing setup. A fastening technique was also employed to connect the optimized SMA plate to the surface of the concrete joint. Finally, some numerical examples have been run in order to check to what extend the utilized method worked properly. The procedure was applied twice; i) when the load combinations were applied in cyclic form and ii) when the load combinations were exerted in reverse cyclic form. Therefore, two optimized SMA have been designed and examined. The results of the analyses showed that the employed technique enhanced the strength of the joint considerably so that the cracking load of the system reinforced with optimized SMA plate under cyclic loading was 1.4 times greater than the benchmark. The load-carrying capacity of the reinforced system in the elastic regime was higher than the unreinforced structure, and the capability in the plastic regime was even higher. Indicatively, the load-carrying capacity of the reference system at a displacement of 32 mm was approximately 98 kN, whereas the respective resistance value was approximately 66 kN in the system without the plate. Besides, the existence of the plate led to transition of the failure zone from the joint to the beam span, which leads to a lower risk of failure of the entire structure. As a result, the main focus of the research was to describe a novel method that allows for a probability-based prediction of damage in concrete structures that can facilitate the assessment and design of degraded structures under risk of failure.enShape memory alloys (SMA)Reinforced concreteColumn-beam jointsProbabilistic analysisOptimizationAnsys690TextMemory-LegierungStahlbetonTrägerWahrscheinlichkeitsrechnungOptimierungANSYSMATLAB