An FPGA-based Real-Time Simulation of a Hybrid Fuel Cell/Battery Source

This paper presents a new approach for the modeling and real-time simulation of an embedded hybrid power source comprised of a fuel cell (FC) and a battery. The proposed modeling is based on a state-space-like representation of the system equations obtained from the modified-augmented nodal analysis. Systematic formulation of system equations is presented and the solution to the non-linear equations are discussed. The proposed model was implemented on an entry level Field Programmable Gate Array (FPGA), and demonstrates sub-microsecond simulation time-step capability. The real-time solution is obtained by precomputing the system equations for all switch state combinations, and using the backward Euler integration scheme for solving differential equations. A fixed point number representation is utilized for speed and reduced configurable resource utilization. Our results show a high fidelity of the proposed model over a wide range of simulation time-steps and switching frequencies.

I. Introduction

With more than 20% of the harmful emissions linked to internal combustion engines (ICEs) [1], automotive manufacturers are working on the electrification of cars, trucks, and buses. Electric Vehicles (EVs), Hybrid Electric Vehicles (HEVs), and Plug-in Hybrid Electric Vehicles (PHEVs) are the alternatives considered to conventional ICE vehicles. Using Fuel Cells (FCs) as the primary source of energy for these vehicles presents an interesting avenue in regard to their effectiveness, high autonomy, and null polluting emissions [2].