Aging-aware real-time multi-layer co-optimization approach for hybrid vehicles: across configuration, parameters, and control

Abstract

The powertrain configuration, parameters, and control of hybrid vehicles are intertwined. All of these factors have a significant impact on acceleration, fuel economy, and battery degradation. However, research on the impact of the multi-layer co-optimization of the powertrain physical configuration and control on battery life has been neglected. Therefore, to fill this gap, an innovative aging-aware real-time multi-layer co-optimization method is proposed in this study. In the topology layer, a novel and improved multi-mode multi-gear configuration is proposed, and the performance differences and the intrinsic mechanism of different powertrain types are analyzed quantitatively and qualitatively. In the control layer, an advanced aging-aware fast real-time control strategy (AFRCS) is proposed. In the AFRCS, the offline optimization layer works in combination with Pareto optimization, battery life aging optimization (BLAO), and parallel computation to speed up the computational efficiency. The online mode coordination layer is used for real-time control, which improves the computational efficiency by approximately 20,000 times, and the battery life optimization is in the range of 10.22 %–12.9 %. During real-world driving cycles, the proposed configuration improves the acceleration and the battery life by an average of 50 % and 22.67 %, respectively, compared to a Toyota Prius, and it improves fuel saving by 8.82 % compared to a Honda Accord. Finally, the proposed AFRCS is verified with a hardware-in-the-loop (HIL) experiment. This study provides guidance for the selection, optimization, and real-time control of next-generation electrified transmissions.