Development of algorithms to identify non-stationary structural systems using statistical filter theories

The purpose of this study is to apply the Monte Carlo Filter for identifying dynamic parameters of a structural system and improve its algorithm. Identifying process using the Monte Carlo Filter sometimes becomes unstable, especially when a parameter has a weak effect on structural responses. In order to overcome this problem, we developed a hybrid algorithm combining the Kalman Filter and the Monte Carlo Filter. A study to investigate the relationships between the Genetic Algorithm and the Monte Carlo Filter has been conducted. Based on this study, we developed an algorithm taking into account the Genetic Algorithm to speed up convergence of the Monte Carlo Filter to identify non-stationary structural systems.