Abstract:
We consider the problem of structural-parametric identification for stationary linear dynamical systems. We propose a new approach for the structural-parametric identification of a dynamical system based on the construction of many weakly equivalent systems of increasing complexity. Structural-parametric identification refers to an estimate of the order of differential equations, all coefficients that satisfy certain constraints, unknown initial conditions, and the measurement bias. The solution to the structural-parametric identification problem is obtained in a finite number of steps. On a test example, we show that the proposed method has high sensitivity even under high-intensity measurement errors.
Keywords:structural identification, parametric identification, linear dynamical systems.
Fulltext:
PDF file (896 kB)
