Low-Speed Online Inertia Identification Considering Inner Disturbances for Dual-Stator Flux Modulated Machine Based on High-Frequency Signal Injection

The article proposes a rotational inertia identification method based on observation object reconstruction and featuring low-fluctuation reference speed for dual-stator flux modulated machine with multiple-PM excitation (DS-FMMPMM) based on high-frequency signal injection. Conventional adaptive observer requires significant reference speed fluctuations and are limited in speed-constrained scenarios. Furthermore, the influence of internal motor disturbances on the accuracy of rotational inertia identification is not considered, especially the harmonic torque disturbances caused by the flux modulation effect in the DS-FMMPMM. The contribution of this article focuses on the theoretical analysis of the impact of internal motor disturbances on rotational inertia identification. High-frequency signal injection is adopted to reduce the influence on the operating state and minimize the influence of disturbance torque on the identification results. The corresponding observation objects are constructed for rotational inertia identification. The rotational inertia adaptive observer can be easily switched with the adaptive disturbance observer. The superiority and effectiveness of the proposed inertia identification approach are validated through simulation and experimental results.