Application of the Extended Kalman Filter—WGI Method in Dynamic System Identification
Masaru HOSHIYA · 2020
In order to obtain stable solutions as well as their fast convergency to the optimal ones in dynamic system identification problems, a method of incorporating a weighted global iteration procedure with the extended Kalman filter (the EK—WGI method) was developed by the author and his co-researchers, and was clarified to be a powerful tool in extensive engineering problems. This paper intends to emphasize the latent usefulness of the method for a case in which the solutions may diverge due to insufficient observation data unless this method is used. Special attention is focused on parameter identification of linear and nonlinear structural systems.