| Autor: |
Cao, Li Ting, Jin, Qi Bing, Fan, Tong Shun, Su, Wei |
| Zdroj: |
Advanced Materials Research; November 2011, Vol. 403 Issue: 1 p3216-3219, 4p |
| Abstrakt: |
On-line identification problem of process model was discussed in this paper, which use warm intelligent technology. An on-line identification method based on HPSO-Rosenbrock parameter estimation algorithm is proposed to solve the problem that traditional identification methods cannot be used in continuous-time systems on closed-loop step response conditions. This identification method is a combined method of a modified PSO and Rosenbrock which can make full use of global search ability of PSO and local search ability of Rosenbrock. Identification results of HPSO-Rosenbrock algorithm were made and compared with the other identification methods. The simulation and compare results show that the on-line identification method proposed in this paper is an approximate unbiased and effective identification method. This method can be successfully applied to closed-loop identification under secious noise and big dead-time object which provides a new idea for system optimization and advanced control. |
| Databáze: |
Supplemental Index |
| Externí odkaz: |
|
