Collaborative optimization of fuel economy and battery life for plug-in hybrid electric buses considering traffic condition

Abstract

The fuel economy of plug-in hybrid electric vehicles is highly related to the supervisory control strategies. This paper proposes a collaborative optimization of energy management strategy considering traffic conditions for plug-in hybrid electric buses. It aims at minimizing energy consumption and battery wear simultaneously while satisfying the driver's power demands and system's constraints. First, an online simplified traffic condition recognition model based on kernel density estimation and weighted naïve Bayesian method is designed to obtain prior knowledge for future driving conditions. Second, a semi-empirical battery aging model is introduced to characterize the battery cycle aging process and incorporated into the control framework. Further, an adaptive equivalent consumption minimization strategy concerning varying traffic conditions and battery degradation is designed, where a back-propagation neural network is formulated to update the equivalent factors online. Finally, the proposed approach is verified under different driving cycles to confirm its superiority over other methods in the field of fuel economy improvement, as well as the postponement of battery degradation. Compared with the charge-depleting and charge-sustaining method, the total driving cost and battery capacity loss of the proposed approach can be decreased by 24.1 % and 18.7 % under the actual driving cycle, respectively.