Abstract

<div class="section abstract"><div class="htmlview paragraph">In the mobility industry, Fuel Cell Electric Vehicle (FCEV) combines fuel cell technology with batteries, allowing to overcome some limitations of Battery Electric Vehicles (BEVs), such as the high recharging time and the high battery mass for applications requiring a high amount of energy (e.g., bus and heavy-duty vehicles). FCEVs have the possibility to collect several information from Intelligent Transportation Systems (ITSs) with relevant potential for vehicle efficiency improvement. Indeed, an online vehicle speed prediction inherently considering real-life factors such as traffic flow, driving path and driving style, allows for novel designs of Energy Management Systems (EMSs) for the optimal operations of the hybrid propulsion system.</div><div class="htmlview paragraph">In such context, this paper proposes an EMS based on Stochastic Model Predictive Control with Learning (SMPCL) to optimize the hydrogen consumption of a Fuel Cell Electric Vehicle (FCEV), while guaranteeing the fulfillment of constraints on battery state of charge (SOC) and available power ranges, as well as maximizing the lifetime of fuel cell and battery. The proposed approach combines a scenario-based Stochastic Model Predictive Control (SMPC) for the propulsion system optimization with novel fuzzy Markov Chains (MCs) for short-term vehicle speed prediction. The effectiveness of this approach has been evaluated considering real driving speed acquisitions of a city bus operating in Turin (Italy) in different traffic flows and with different drivers. For comparison, several algorithms have been applied to a high-fidelity simulation plant representative of the FCEV propulsion system developed in GT-SUITE. The results show that SMPCL allows for relevant reduction of hydrogen consumption compared to classic rule-based approach, while getting also important benefits in terms of fuel cell and battery lifetimes. Moreover, hydrogen consumption is very close to the results of a global offline optimization algorithm used as benchmark (i.e., Dynamic Programming). Finally, next steps will include experimental validation of proposed approach on a real propulsion system in a test bench located in Turin (Italy).</div></div>

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.