[EN] This study proposes a predictive equivalent consumption minimization strategy (P-ECMS) that utilizes velocity prediction and considers various dynamic constraints to mitigate fuel cell degradation assessed using a dedicated sub -model. The objective is to reduce fuel consumption in real -world conditions without prior knowledge of the driving mission. The P-ECMS incorporates a velocity prediction layer into the Energy Management System. Comparative evaluations with a conventional adaptive-ECMS (A-ECMS), a standard ECMS with a well -tuned constant equivalence factor, and a rule -based strategy (RBS) are conducted across two driving cycles and three fuel cell dynamic restrictions (|di/dt|max <= 0.1, 0.01, and 0.001 A/cm2s). The proposed strategy achieves H2 consumption reductions ranging from 1.4% to 3.0% compared to A-ECMS, and fuel consumption reductions of up to 6.1% when compared to RBS. Increasing dynamic limitations lead to increased H2 consumption and durability by up to 200% for all tested strategies.

This research is part of the project TED2021-131463B-I00 (DI-VERGENT) funded by MCIN/AEI/10.13039/501100011033 and the European Union "NextGenerationEU"/PRTR.