Adaptive multi-objective optimization strategy for real-time energy management of fuel cell vehicle

In this study, we propose an adaptive multi-objective optimization energy management strategy for optimizing the instantaneous power split in fuel-cell/battery electric vehicle. The optimization objectives include improving fuel economy, maintaining the state of charge of battery at a medium level, and extending fuel cell lifespan by suppressing dramatic changes in fuel cell power. The weighting factors of the objective functions can be dynamically adjusted based on realtime operating data of the hybrid power system. The proposed strategy is simulated and compared with a rule-based strategy under four different driving cycles. The results demonstrate that our strategy can reduce fuel consumption in varying degrees, and it allows the fluctuations in state of charge of battery to be always controlled within a reasonable range. The most outstanding aspect of this strategy is that it enables an over 80% reduction on the average variation rate of fuel cell power. Our strategy has the features of not relying on prior knowledge of the driving cycles and precise modeling of the power system, giving it great potential in real-time applications.