Original scientific paper

MODEL UPDATE WITH THE OBSERVER/KALMAN FILTER AND GENETIC ALGORITHM APPROACH

Željan Lozina ; Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, University of Split, Split, Croatia
Damir Sedlar ; Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, University of Split, Split, Croatia
Damir Vučina ; Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, University of Split, Split, Croatia

Abstract

The discrete time Observer/Kalman model identification technique is implemented in order to identify the structure pulse response. The model updating procedure based on the finite element model pulse response of the test structure and the genetic algorithm is developed. The objective function evaluates the difference between the system and the model pulse responses. The modal assurance criteria implementation is considered. The model reduction in order to match the model degrees of freedom (dofs) and the test structure dofs involved in the experiment is discussed. A case study on the frame test structure is provided.

Keywords

system identification; model update; observer; Kalman filter; genetic algorithm

Hrčak ID:

89196

URI

https://hrcak.srce.hr/89196

Publication date:

10.7.2012.

Visits: 1.289 *