Application of optimization procedures using phantom connectivity technique

The objective of this thesis is to identify global structural differences between a finite element model and a target model defined using structural dynamic information. Optimization procedures identify the differences and adjust the system matrices of the model to achieve the target specification. Constraints of the finite element topology may be used in these procedures. In order to improve these optimization processes, a new technique is proposed. This approach (Phantom Connectivity Technique) modifies the existing topology of the finite element model to allow additional elemental connectivities in the optimization process. The use of the Phantom Connectivity Technique allows for the optimization procedures to identify the changes outside the skyline of the original finite element model in the full space model. In applications where no finite element topology exists (such as with reduced component models), the Phantom Connectivity Technique is used to generate a finite element constraint topology; this allows for the optimization procedures to identify changes in the reduced component model.; Several applications using the optimization processes along with the Phantom Connectivity Technique are investigated using component system models with simulated target information; both full and reduced finite element models are used for component definitions. Various numerical examples are used to evaluate the robustness of this proposed technique. The technique is shown to produce very acceptable results for the target model scenarios investigated.