Fuel Cell Vehicle simulation: an approach based on Toyota Mirai

The high expectations on hydrogen in the global energy system of the early 2000s failed to be met. In recent years, because of the more stringent emission legislation framework, the European funding for low emission innovations and the technology improvement combined with cost reduction, hydrogen and Fuel Cells are regaining momentum. In this context, the present thesis work focuses on the modelling of performance and fuel consumption of Fuel Cell vehicles (FCVs). Starting from the models present on Advisor, a MATLAB/Simulink model capable to simulate the whole vehicle system during various driving cycles has been created. The main components are modelled as follows. The vehicle is represented through its load parameters and longitudinal dynamics. The motor is described through a torque-speed map with efficiency curves. The battery is modelled through its main parameters such as the state of charge, the capacity, the voltage and the internal resistance. The whole system is considered for the Fuel Cell. The stack is modelled with its polarization curve while the auxiliaries are modelled with operational maps if present like for example in the air compressor or with their efficiencies in the case of the pumps. Firstly, the data of an average small car are considered. The simulation is run over five different driving cycles, namely FTP, NEDC, UDDS, US06 and WLTP, and different control strategies are evaluated. Thus, a comparative analysis on the collected results is performed. The control strategy which highlighted the best performance in terms of fuel consumption and driving range is selected for the second part of this work. Afterwards, the main parameters of the model, such as the battery, fuel cell stack, electric motor, aerodynamic drag and vehicle parameters, have been set accordingly to those of 2017 Toyota Mirai Fuel Cell Vehicle. The analysis has been carried out considering both the tests made on Toyota Mirai vehicle by Argonne National Lab (ANL) in 2017, and looking at available data relative to similar vehicles. Finally, a model which closely represents the Toyota Mirai vehicle has been built. For validation purposes, the simulation has been run on the NEDC and US06 cycles and the results have been compared to those found by the Argonne National Lab. The thesis work ends with critical comments about the obtained results and the possible future works perspectives.