Iterative Learning Control Design of PMSGs for Hybrid Vehicles

This paper proposes an iterative learning control (ILC) design that significantly improves torque tracking performance in both transient and steady states of a permanent magnet synchronous generators (PMSGs) for the electrification (hybrid) vehicle. The proposed ILC design has the following benefits in two folds: feedback and compensation control (FCC) and iterative learning control (ILC). The former achieves good torque tracking performance by using compensation terms and latter uses stored data (i.e., previous control inputs) to improve the performance of the control inputs. In particular, the ILC design suggests a way to guarantee a good torque tracking performance regardless the PMSG information by rejecting the periodic and non-periodic disturbances. In addition, stability is achieved because the torque tracking error asymptotically reaches to zero. The proposed ILC design provides good transient performance (e.g. fast transient response and small overshoot), steady-state performance (e.g. small steady state error), and robustness to parameter uncertainties. To show the effectiveness for the proposed ILC design, it is implemented in co-simulation by the MATLAB/Simulink and PLECS. And the small-scale HILS (Hardware-in-the-Loop Simulation) system for the PMSG and engine modeling with TI DSP is implemented.