摘要:AbstractSwitched reluctance machine (SRM) is a highly nonlinear system whose modeling is very complex. Indeed, SRM is usually operated in the saturation zone of its magnetic so that energy transfer is maximized. The existing models do generally not account for the hysteresis effect. This work develops a system identification approach to obtain SRM models considering magnetic nonlinearity. The proposed identification approach is frequency geometric nature performing a model search in the class of Wiener models. The latter proves to be convenient for accurately capturing the nonlinear behavior of SRMs. The proposed method involves system excitation using sine inputs. The results obtained with this method are compared with those obtained using the magnetostatic finite element method. The comparison reveals better results for the former featuring better accuracy of flux-linkage value characteristics.
关键词:KeywordsSRMHysteresis effectSaturation effectWiener modelNonlinear system identification