Real Time Simulation of Proton Exchange Membrane Fuel Cell Hybrid Vehicle Christian Dufour 1 , Tuhin K. Das 2 , Shankar Akella 2 1 Opal-RT Technologies Inc. ,1751 Richardson, bureau 2525, Montréal QC, Canada H3K 1G6 Phone. 514-935-2323, christian.dufour@opal-rt.com , http://www.opal-rt.com 2 Emmeskay inc., 44191 Plymonth Oaks Blvd, Plymouth, MI, 48170 Phone: 734-207-5565, tuhin@emmeskay.com , shankar.akella@emmeskay.com Abstract – This paper will present results on the real time and rapid simulation of a Proton Exchange Membrane (PEM) based fuel cell hybrid electric vehicle. The fuel cell hybrid vehicle (FCHV) is composed of a battery, a fuel cell, a DC-DC converter and a PMSM motor drive. The converters in the circuit are modeled with special interpolating IGBT bridge models (RT-Drive Time Stamped Bridges) that are very accurate despites the 10 kHz carrier frequency and the 10 microseconds sample time of simulation. The detailed Emmeskay PEM fuel cell stack model is used for the simulation. In the paper, the RT-LAB system is demonstrated to accelerate simulation by a factor greater than 1000 and help decrease the development cycle time of FCHV. Keywords – real time simulation, hardware-in-the-loop, fuel cell, PEM fuel cell, electric motor drives, interpolation. I. INTRODUCTION The fuel cell car concept constitutes a major leap in the domain of personal transportation. Major problems remains for its wide acceptance like hydrogen infrastructure creation and cost effective hydrogen production to name a few[18]. At the vehicle design level, important effort have been made by some manufacturers to demonstrate real advantage to shift to this new paradigm. Toyota Motor Corp., for example, is currently actively developing a commercial fuel cell hybrid electric vehicle and regularly reports on its advances[2][3], using the latest breakthrough in simulation technology for that purpose[4]. The main promise of fuel cell powered car is a pollution-free high-efficiency car with good performance. While nobody doubts that a power source with water exhaust is environmentally friendlier that an internal combustion engine, the performance issue is more complex. For example, the proton exchange membrane fuel cell has a relatively slow dynamics for car applications and therefore a temporary power storage device, like a battery or super-capacitor, is required to maintain performance during peak power demands. The same storage device also helps on the efficiency issue by recuperating power during vehicle deceleration. The whole fuel cell hybrid vehicle design with fuel cell, battery, converters and power train is somewhat a complex apparatus and advanced techniques are required to test it effectively. Lack of prior experience, expensive equipment and shorter developmental cycles are forcing researchers to use model based analysis techniques for development of the fuel cell control systems. Model based control system development depends heavily on the availability of accurate mathematical model of the different FCHV subsystems, the fuel cell being one important one. There are several PEM fuel cell models available in literature[12][13][14][15] but most of these models are design-oriented for the purpose of component selection or operating point optimisation. They are not suitable for the control development, which requires models that can predict the behaviour on the fly when an input variable such as fuel flow rate is changed. A good example of detailed dynamic fuel cell model in given in[10][11] which include anode and cathode mass flows, hydration and stack voltage modelisation but without thermal effects. Development and testing such a complex system is usually made at several levels starting from individual subsystem dimensioning and control, and finalising with overall system optimisation and control in standard and HIL simulation. It happens that the late stages of optimisation and test can be very time-consuming because of the size of the model combined by the simulation technique used. 1