Automated SOFC Design Exploration

摘要

We present automated computer aided engineering (CAE) to maximize power efficiency and minimize temperature variations in fuel cell stack design. Validated cell models are applied to preliminary 3-D stack simulations, where proprietary optimization algorithms quickly obtain better multi-parameter designs. Each simulation conserves flow, energy, species, and charge transport using segregated solvers with a high degree of linearization and efficient sparse matrix solvers. Automated workflows and state-of-the-art search algorithms allow for fast design space exploration of fuel cell dimensions, properties and operating conditions. Efficiency relates electric power output to reactant enthalpy loss and air-compression work input. To minimize temperature variation, while maintaining high efficiency for fixed fuel flowrate of 7.35E-7 kg/s, an optimal air flowrate of 2E-4 kg/s is preferred for square cells of 36 cm~2 active area and 5.25 mm height. Wide serpentine channels with cathode supported membrane-electrode-assemblies (MEA) minimize pressure drop and diffusion length scales. Serpentine channels entering and leaving in opposite corners of the cell are found preferable for minimizing temperature variations.