Simulation of the Fuel Cell Thermal Behavior with Unified Parallel C

In this paper, a parallel simulation to analyze the thermal behavior of proton exchange membrane fuel cells (PEMFC), using Unified Parallel C (UPC) is presented. PEMFCs are considered as very promising alternative solution for energy generations in the future for stationary and mobile applications. In particular, they are the most favored fuel cell (FC) type to power electric vehicles of the future as a replacement for the internal combustion engine. The computation model based on the global nodal network method (GNNM) used to describe the thermal behavior presents high scalability and intrinsic parallel characteristics suitable for parallel implementation. The objective of this work is to develop a parallel implementation of GNNM using the Unified Parallel C (UPC) for a cluster of processors according to the distributed shared memory programming paradigm. UPC is an explicit parallel language that provides the facilities for direct user specification of program parallelism and control of data distribution and access to maximize performance. Due to its global address space, UPC simplify the management of ghost zones in large-scale simulations. Simulations results shows that computation time to conduct thermal behavior simulation is drastically reduced compared to sequential simulations and obtained simulation data are highly accurate and precise.