The Hybridized Fuel Cell Vehicle Model of the University of California, Davis 2001-01-0543
Vehicle manufacturers claim that fuel cell vehicles are significantly more fuel-efficient and emit fewer emissions than conventional internal combustion engine vehicles /1/. A computer model can help to explore and understand the underlying reasons for this potential improvement. In previous published work, the UC Davis Vehicle Model for the case of a load-following Indirect Methanol Fuel Cell Vehicle (IMFCV) has been introduced and discussed in detail /2/.
Because of possible technical barriers with load following vehicles, as well as near term cost issues, hybrid fuel cell vehicle concepts are widely discussed as another fuel cell vehicle option. For load following vehicles, the questions of fast start up and fuel processor dynamics in extreme transient situations, (e.g., during phases of hard acceleration) are not totally resolved at this time. For both of these performance issues, a hybrid design could offer at least an interim solution.
To investigate the potential of the Indirect Methanol Fuel Cell Hybrid Vehicles (IMFCHV) further, the original model for the load following IMFCV has been expanded. First, a battery component model has been added (including a model of a battery controller). Second, a dc-dc converter component model has been included between the fuel cell stack and the battery (to ensure the proper integration of the battery into the overall operation of the electrical system).
The purpose of this work is to explain the underlying modeling philosophy, as well as the algorithms for the hybrid case. The approach followed in this IMFCHV model allows parametric studies to be made over a large range of configurations, and simplifies the effort involved in deciding the optimal design for a given scenario.
Several examples are given to illustrate the capabilities of the model. However the paper is not meant to exhaustively answer specific questions about fuel economy or other vehicle properties. Instead the focus is primarily on the explanation of the algorithms of the simulation, in order to allow a better understanding of the model itself.