Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
1714055 | Nonlinear Analysis: Hybrid Systems | 2007 | 16 Pages |
Abstract
We consider the identification of a switched linear system which consists of linear sub-models, with a rule that orchestrates the switching mechanism between the sub-models. Taking a set of switched linear systems and using a state space framework, we show that it is possible to combine subspace methods with mixed integer programming for system identification. The states of the system are first extracted from input–output data using sub-space methods. Once the state variables are known, the switched system is re-written as a mixed logical dynamical (MLD) system and the model parameters are calculated for via mixed integer programming. We report an example at the end of this paper together with simulation results in the presence of noise.
Related Topics
Physical Sciences and Engineering
Engineering
Control and Systems Engineering
Authors
P. Egbunonu, M. Guay,