Nonlinear deadbeat-direct torque and flux control for highly saturated synchronous reluctance machines in automotive traction applications

This research extends prior research by presenting nonlinear deadbeat-direct torque and flux control (DB-DTFC) for highly saturated synchronous reluctance machines. Main objective is the improvement of the controls performance for automotive traction applications. Flux linkage including harmonics is observed with a Gopinath observer and precisely closed-loop controlled with high bandwidth by applying DB-DTFC. Nonlinear DB-DTFC is developed by incorporating saturation and cross-saturation directly into the differential torque equation. Flux linkage-based loss models for machine and inverter are evaluated experimentally. Loss model-based flux magnitude is calculated in real-time and overall drive losses of SynRMs are minimized each single switching period. The excellent flux weakening capabilities of SynRMs combined with the direct control of flux applying DBDTFC lead to a measured efficiency improvement of 1.4% under the highway driving cycle HWFET. Increased disturbance rejection, as compared to current vector control (CVC), is achieved via simulation and experiment. Minimization of harmonics in current and a reduction of total loss are measured on a 5.5kW synchronous reluctance machine (SynRM) testbench. Moreover, real-time flux observer-based torque ripple estimation is presented. When flux estimation is accurate, instantaneous torque is fed back and unwanted pulsating torque is inherently minimized. FEM-simulations validate the torque ripple minimization.