Abstract
Intelligent approaches were applied to the procedure of nonlinear modeling for switched reluctance motor (SRM) and exact estimation of the rotor position. From the measured data of stator current, winding inductance and rotor angle position in experiment, the nonlinear characteristic of SRM was obtained and fuzzy rules were further developed. Learning ability of neural network and global optimization of genetic algorithm (GA) was employed to minimize the modeling error. The proposed algorithm can realize exact estimation of the rotor position with certain robustness and reliability and can directly replace the sensor of position. Complexity can be avoided by intelligent modeling of nonlinear system. The results of simulation demonstrate efficiency of the modeling.
Original language | English |
---|---|
Pages (from-to) | 722-729 |
Number of pages | 8 |
Journal | Tianjin Daxue Xuebao (Ziran Kexue yu Gongcheng Jishu Ban)/Journal of Tianjin University Science and Technology |
Volume | 38 |
Issue number | 8 |
Publication status | Published - Aug 2005 |
Externally published | Yes |
Keywords
- Fuzzy logic
- Genetic algorithm
- Neuro-networks
- Switched reluctance motor (SRM)
ASJC Scopus subject areas
- General