Vision-Based Structural FE Model Updating Using Genetic Algorithm

Structural members can be damaged from earthquakes or deterioration. The finite element (FE) model of a structure should updated to reflect the damage conditions. If stiffness reduction is ignored, analysis results will unreliable. Conventional FE updating techniques measure response with accelerometers update model. However, only where sensor installed. This paper introduces new computer-vision based method for structural using genetic algorithm. system measures displacement seven different object tracking algorithms, and optimizes parameters To validate performance, lab-scale test three-story building was conducted. each story measured before after reducing one column. Genetic algorithm automatically optimized non-damaged state state. proposed successfully expected reduce time cost updating.