This paper presents an adaptive energy management strategy (EMS) based on Pontryagin's Minimum Principle (PMP), considering both historical driving cycles and stochastic vehicle mass. First, an improved battery state of charge (SOC) feedback control method based on Proportion-Integration (PI) controller is proposed for adaptive energy management. Second, a Taguchi robust design (TRD) method is proposed by considering the noises of historical driving cycles and stochastic vehicle mass. The main purpose of the TRD is to find the robust parameters (P and I) that can counter to the above noise disturbance, thereby neglecting the influence of noise disturbance in adaptive energy management control and minimizing the mean hydrogen consumption. The results show that the TRD is reasonable and applicable in the robust design of plug-in fuel cell electric buses (PFCEBs); the fuel economy of the proposed strategy can be averagely improved by 18.36%, compared to the rule-based energy management strategy.