Novel Segmented-Prediction-Based FCS-MPCC for Low-Control-Frequency EV EESMs with Uncertain Mutual Inductance Considered

Electrically excited synchronous motors (EESMs) without installing slip rings and brushes are drawing increasing attention in the electric vehicle (EV) propulsion systems. To improve the control performance of the EV EESMs with uncertain mutual inductance, which works under low control frequency (LCF), this paper proposes a novel segmented-prediction-based finite control set model predictive current control (FCS-MPCC) strategy. First, a sliding mode (SM) observer is constructed to identify the mutual inductance, with its stability and robustness against parameter mismatch analyzed. By using the estimated mutual inductance, the accurate EESM model used for FCS-MPCC is established, Second, the segmented prediction algorithms are developed to reduce the prediction errors caused by local linearization in the LPF situations. Finally, the proposed mutual inductance identification and high-performance control techniques are verified by experiment, which is conducted on a 580-W EESM drive system.