Optimization of Steel Exoskeletons for the Seismic Retrofit of Reinforced Concrete Structures via Genetic Programming

In the last decade, the use of steel exoskeletons as an alternative to seismic retrofit of existing structures received great attention due to their non-invasive and time-efficient nature. Furthermore, it represents the preferred solution in cases where the use of the existing building cannot be interrupted during the intervention, or when a limited level of knowledge of the existing building is available and high seismic activity is expected. Nowadays, the proposed design approaches mainly rely on simplified models employing single or multi-degree-of-freedom systems with lumped masses to determine optimal overall values for exoskeletons' stiffness, mass and/or damping. Even though the benefits derived by these methodologies were largely proven, further investigations are required to fully understand the influence of mechanical and geometrical properties on the outcomes. Moreover, once global optimal values of the system have been obtained, no consolidated procedures for the sizing process of the exoskeleton have been recognized in the literature. Therefore, the primary objective of this research is to overcome the limitations associated with existing methodologies by evaluating the optimal quantity and spatial placement of the exoskeletons, as well as determining the optimal sizing of their constituent elements. To accomplish this goal, a comprehensive optimization procedure, based on a modified Genetic Algorithm has been adopted. In the mathematical statement of the problem, the total weight of the exoskeletons has been assumed as the Objective Function (OF). Two critical constraints are considered: firstly, aiming to maintain the structural integrity of the building in the elastic range, a maximum allowable inter-storey drift has been imposed corresponding to the threshold beyond which damage occurs at the level of the structure’s infills. Secondly, the stress requirements that the exoskeletons must satisfy have been considered. To determine the fitness and assess the viability of each potential solution, modal analyses are conducted using SAP2000 OAPI, a software tool that facilitates the generation of models with automatic routines and allows the implementation of optimization tools governed by MATLAB. The results obtained from this research demonstrate that the proposed approach effectively controls damage in existing structures subjected to horizontal actions, ensuring compliance with safety requirements while offering slender and cost-effective designs. Additionally, the stiffness ratio between the exoskeletons and the base structure can be significantly reduced below the limit suggested by the NTC2018 Italian code. Comparative analyses of different exoskeleton configurations and base structures provide valuable insights into their performance, enabling a better understanding of their behavior and aiding in the selection of appropriate retrofitting strategies. The thesis presents the following organization: in Chapter 1 the topic of the thesis is introduced together with the definition of the scope and main goals of the work, in Chapter 2 a well-comprehensive literature review is reported by summarizing the main results achieved by other authors. The definition of several case studies, the F.E. modelling and the problem statement of the optimization are presented in Chapter 3. In Chapter 4 the results for each scenario are discussed while the final conclusions and the potential future developments of the work are explained in Chapters 5 and 6, respectively.