Energy Management of P2 Hybrid Electric Vehicle Based on Event-Triggered Nonlinear Model Predictive Control and Deep Q Network

Hybrid electric vehicles (HEVs) are used as a bridge during the transition to battery (BEVs) and make energy consumption more efficient. The main problem in improving efficiency of HEV is torque management. In this study, novel approach based on nonlinear model predictive controller solve reference tracking distribution proposed. That say, order increase distribution, weights control (NMPC) trained with Deep Q Network (DQN), an event-triggered mechanism designed DQN reduce computational cost MPC. considered varies according type structure HEV. parallel 2 hybrid vehicle (P2 HEV) modeled via publicly shared passenger data engine, motor, high-voltage battery, transmission, clutch, differential, wheel characteristics. NMPC formulated so that values remain within physical limits also operates at its limits. Namely, it guaranteed works certain state charge (SOC) window current health (SOH) optimization. motor engine efficiencies by 3.61% 2.86%, respectively, proposed structure, while reduced 52.01% when utilizing event-triggering controller.