Dual-layer approach for systematic sizing and online energy management of fuel cell hybrid vehicles

Abstract Numerous shortcomings such as higher manufacturing cost, inadequate refueling facilities, and durability issues are impeding the competitiveness of fuel cell hybrid vehicles (FCHVs). Effectively coordinating the design and operational aspects with a fair trade-off among ownership price, energy consumption, and health degradation is the key to address these shortcomings. Against this background, the paper devises a novel dual-layer approach to synergize the sizing aspect with the online energy management system (OEMS) of FCHV. Within the proposed approach, the first layer utilizes the specific driving parameters to offer a global configuration of the FCHV propulsion system. Despite guaranteeing consistent vehicular performance under diverse driving conditions, the global configuration generally favors oversized sources. To address this problem, flexible design coefficients are introduced to allocate a practical sizing range to the sources of FCHV. From this range, a feasible combination is selected in the second layer by coordinating the sizing procedure with the OEMS. An effective coupling method is exploited to design the OEMS, where two distinct supervisors with instantaneously adjustable rules are formulated. The primary supervisor comprises a search-horizon based online optimizer to tune the frequency-driven power splitting rules, subject to fuel economy, health degradation, and storage violation. Subsequently, a droop-adjustment procedure is integrated as secondary supervisor to realize the storage limitations. An in-depth qualitative analysis yields a global propulsion configuration with a suitable compromise among ownership price, fuel economy, and health degradation, highlighting the efficacy of the proposed approach. At the OEMS level, compared to competing variants of frequency-driven power splitting methods, the expected improvements in fuel consumption and health degradation costs can approximately reach 14% and 29% in the studied driving environments.