Initial value assumption for Estimation of Structural Dynamic System using Extended Kalman Filtering

구조물의 동특성치 예측을 위한 확장칼만필터기법의 초기치 설정에 관한 연구

Extended Kalman Filter iterate the prediction and the filtering based on Initial state for the next time step. EKF method for the estimation of nonlinear parameters of a structural dynamic system is necessary that initial of state vector and error covariance matrix. Because those are unknown exactly, generally selected random values. That occasion observability problem appear because of unknown initial values. In this study, for the estimation of the nonlinear parameters, a simple one degree of Freedom example is carried out by Extended Kalman Filter. And initial value assumption for Parameter Estimation of Dynamic System are developed. The result of analysis is compared with calculated standard values.