First, a control FCHEV powertrain model is developed to design power split control strategies.
The next section describes the FCHEV control model.
A FCHEV control model including Longitudinal Vehicle Dynamics (LVD), FC and battery submodels is developed in a MATLAB/Simulink[R] environment.
Total FCHEV energy includes the energy consumed in the battery in addition to the energy used in the FC which can be described by available energy to be extracted from the amount of hydrogen.
where [E.sub.tot] is total consumed energy on the FCHEV, [E.sub.batt] is the battery consumed or saved energy, [mathematical expression not reproducible] is mass of consumed hydrogen, and [mathematical expression not reproducible] is hydrogen lower heating value.
Table 1 lists the vehicle, FC, and battery parameters for the FCHEV control model.
Power flow and interactions between supervisory controller and FCHEV powertrain components of this study are illustrated in Figure 3.
For FCHEV power management control, moving backward (recursively) from the end of the driving cycle, it is needed to calculate the optimal cost function at each time step.
Then, at each time step, the minimum of the costto-go functions in that column of the matrix is selected as the optimal cost-to-go function at that time and the corresponding current split ratio is saved as the appropriate control input to the FCHEV system.
For the studied FCHEV power split control, different prediction horizon lengths are tested for comparison purposes while the simulation step time is fixed and equals 1 s.
Urban Dynamometer Driving Schedule (UDDS) drive cycle is used to design and test the FCHEV power split control strategies.
Table 2 lists the values of total energy consumption and battery SOC for the DP and MPCs from the FCHEV simulation.