Font Size: 

Series hybrid electric vehicles (HEVs) allow engine operating points to shift into higherefficiency regions by leveraging the hybrid energy system. Usually, standard energy management systemsapproximate energy conversion dynamics using steady-state maps, which leads to a sub-optimal controlpolicy when fuel consumption and pollutant emissions are observed under engine transient demands. Toovercome this issue, this paper proposes a multi-objective optimization strategy to reduce these indices. AGrey Wolf Optimization (GWO) metaheuristic algorithm is used to optimize the Nonlinear Model PredictiveController (NMPC) parameters offline using a MATLAB/Simulink model built from the experimentalmean-value maps of an ethanol-fueled internal combustion engine (ICE) with transient considerations. Finally, the parameterized NMPC is used to experimentally control an ethanol genset across five powertransitions. The adaptive approach reduced the air-fuel ratio error (˜λ = 1) by an average of 8.56% andfuel consumption by 4.3% when compared to conventional PI-based controllers.