In order to improve the accuracy of driving pattern identification in the adaptive energy management strategy of extended-range electric vehicle (EREV), the effective characteristic parameters were selected by analyzing the correlation between the sample characteristic parameters and the optimal co-state variables of Pontryagin's minimum principle (PMP). The sample database was established after the principal component analysis of effective characteristic parameters. In order to realize the adaptive adjustment of PMP energy management strategy, the on-line driving pattern identification was carried out, and the identified optimal co-state variable was taken as the current co-state variable after the correction of battery SOC. The results show that the fuel consumption of the designed adaptive PMP energy management strategy is lower than that of the current driving pattern identification strategy. The SOC change is more stable, which can extend the life of power battery.
