Skip to main content
Research Project

Solid oxide fuel cell modelling from cell to system

Multi-scale and dynamic tank-in-series models of solid oxide fuel cells, plus observer design to estimate internal states, built for process simulators.
Solid oxide fuel cell modelling from cell to system

Why it matters

Solid oxide fuel cells convert fuel directly to electricity with high efficiency and low emissions, and can run on fuels such as hydrogen, biogas or reformed hydrocarbons. What holds them back is durability. Temperature gradients and uneven fuel supply inside a cell or stack shorten its life, and experiments on stacks are expensive.

The problem

Commercial process simulators have no built-in SOFC module, so a fuel cell is usually treated as a lumped black box. That hides the internal conditions that decide whether a stack lasts: thermal hotspots, local fuel starvation and slow transients. Full CFD can resolve them, but it is too costly to use for system studies or control.

What we do

  1. Multi-scale modelling. Build a rigorous SOFC module inside a flowsheet simulator, from a single compartment up through channel, cell, stack and system, so that stack behaviour and the surrounding plant can be studied together.
  2. Fast dynamic models. Represent a direct internal reforming cell as tanks in series. This keeps the spatial and transient behaviour at a small fraction of the computing cost of CFD.
  3. State estimation. Design an observer that estimates temperatures, species concentrations and unmeasured inputs from limited measurements.

Approach

The models are checked against experimental data, and CFD is used for spatial profiles. The stack-scale studies look at flow maldistribution in the manifold and at recycling depleted anode gas, which can be used to manage gradients and improve fuel utilisation and water management. The tank-in-series model reproduced experimental voltage-current curves to within about 1.5%.

Outputs

Collaborators

Curtin University and Ceramic Fuel Cells Ltd.

Contact

For collaboration or student projects in fuel cell and reactor modelling, contact the SMILE lab.