Degradation adaptive energy management with a recognition-prediction method and lifetime competition-cooperation control for fuel cell hybrid bus

Auxiliary power sources such as batteries and supercapacitors are commonly used in fuel cell buses to meet complex power requirements and extend the life of the fuel cell. However, different power supplies have different sensitivity to the actual driving conditions which may cause degradation imbalance, reducing the service lifetime of the whole hybrid system. To solve the problem, a lifetime game optimization management strategy for fuel cell triple-source hybrid bus buses is developed in this paper, which takes into account the competitive relationship between fuel cells and batteries. To begin, a method for predicting driving cycles based on learning vector quantization (LVQ) and back-propagation (BP) neural networks is presented as to improve forecast accuracy. Second, the relational expression of hydrogen efficiency decreasing with the fuel cell state-of-health (SOH) is further derived, forming a new fuel cell degradation model that takes different decay rates of the fuel cell under different operating conditions into account. Finally, the competition-cooperation mechanism between fuel cell and battery is described as a double-source lifetime degradation game optimization strategy based on non-cooperative game theory. In the game optimization process, the latest SOH is used to calculate the hydrogen efficiency decline in real-time, so that the strategy can obtain the degradation adaptive property. The simulation results show that the presented strategy improves the economy by 81.64% compared with the rule-based strategy. Compared with the traditional model predictive control energy management strategy, the economic cost is reduced by 76.99%, while the degradation of the fuel cell is reduced by 76.83% at the cost of 49.28% cell degradation elevation.