Applying Genetic Algorithms on Behavior analysis of Damaged Structures

• Main Authors: Zhong-JeiLiu, 劉中杰
• Other Authors: Deh-Shiu Hsu
• Format: Others
• Language: zh-TW
• Published: 2010
• Online Access: http://ndltd.ncl.edu.tw/handle/50571919689852165936

博士 === 國立成功大學 === 土木工程學系碩博士班 === 98 === The research describes the non-destructive damaged structures using the acceleration sensor measurement data for four-story shear structures, and obtains the optimal solution of story stiffness by genetic algorithm. The fitness function, based on the similarity of measurement-curve derives from the acceleration sensor with responses of the considered structure to find the optimal gene and the stiffness of the damaged structure. The research aims a four-story shear typed structure, including cases with incomplete acceleration sensors. Cases with measurement data noises are also discussed. The simulation results identify the stiffness of damaged-stories using fewer displacement sensors if the sensor location is properly arranged. The genetic algorithm proves that neglecting the noise of incomplete measuring data from other sensing points can still successfully obtains the optimal solution. The current study discusses the experiment case of the four-story reinforced concrete structure, and finds the stiffness reduction by the genetic algorithm. Fluid dampers have received considerable attention in recent years because of their low cost and usually will not induce additional here in structural systems. However, analysis of structures with fluid damper devices reveals that we are facing a nonlinear dynamic problem because of the relationship between force and velocity of the damper is highly nonlinear, although the structure behaves linearly, the whole damper-structure system has inherent nonlinear properties. The stochastic linearization technique (SLT) is typically performed with a linear system that is statistically equivalent to the nonlinear one. In this paper, an alternative technique, a genetic algorithm (GA), is applied, which is the most used optimal searching method in recent years. The motion equation of the structures is utilized as the objective function. And the non-destructive detection of structure with fluid damper devices are also detected herein.