Fuel cell electric vehicles (FCEVs) are gaining increasing attention as a potential solution to reduce greenhouse gas emissions and improve energy efficiency in the transportation sector, because they are generally seen as offering higher range and faster refuelling times compared to pure battery electric vehicles. However, the performance and consumption of fuel cell vehicles heavily depends on the powerplant component sizing, the energy management and fuel-cell mode operation strategy, as well as the driving conditions. To efficiently optimize the energy management strategies of FCEVs under a wide range of driving conditions, the availability of simulation tools become essential, and model-based calibration becomes a suitable approach to shift development and validation tasks from a physical testing approach into a virtual simulation environment, allowing for increased system robustness while also reducing development costs and time. Due to the complex system interactions and different mode operations of a FCEV, a dynamic model with a high degree of accuracy of the integrated vehicle-powertrain system is required for system-level optimization and fast development of the necessary energy management control and FC command system. A commercially available FCEV, in this case a Hyundai NEXO, was selected to be measured on the chassis dynamometer facility, with the goal of measuring power consumption of each component, to extract the characteristics of the system and derive an equivalent virtual vehicle model of a FCEV, that is used to draw an exhaustive energy balance of the vehicle powertrain on various mission profiles. This paper presents a methodology to generate a reliable virtual model from experimental data to virtually assess the impact of control and component variations on vehicle performances and energy consumption. The proposed framework incorporates a detailed model of a proton exchange membrane (PEM) fuel cell system, including the fuel cell stack and the balance of plant (BoP) components (air compressor, humidifier, cooling system, etc), a high voltage battery, a DC/DC converter, an electric motor/generator, and vehicle model as the most important parts. Additionally, to be able to follow a transient drive cycle is also necessary to create the required control loops (for FC and battery control) with an energy management strategy and a driver model. This model development methodology based on test measurement is divided in three consecutive stages: parameter identification, model calibration and model validation on a different set of experimental data than the used for calibration. After completing the proposed model development methodology, the final error in predicted hydrogen consumption in WLTC cycles remains below 3%, and below 1% when comparing the accumulated e-motor energy consumption. The resulting model runs faster than real time, which enables efficient optimization of the energy management control strategy on longer mission profiles such as RDE (real driving emission routes). The simulation framework is also used to evaluate the energy consumption of the FCEV under various driving scenarios, including urban, suburban, and highway driving, and to perform a complete energy breakdown analysis across the studied conditions. The results show that the energy consumption of the FCEV is highly dependent on the driving behaviour, with significant differences observed between aggressive and eco-friendly driving styles. The simulation results also demonstrate the potential benefits of using a slightly larger battery pack in conjunction with the fuel cell system to allow a higher pure-electric range and more freedom to better manage the SOC of the battery depending on the driving conditions to improve the overall energy efficiency of the vehicle. Keywords: Fuel-Cell Electric Vehicles, FCEV, virtual simulation platform, energy consumption analysis

Ing. Marina Roche, Project Manager Virtual Development, Applus IDIADA