Reliable Fuel Cell Simulation Using an Experimentally Driven Numerical Model

The necessity of a simulation tool for integrated power plant with MCFCs has led to the development of a numerical model by which it is possible to evaluate the performances of the cell and the plant the cell is embedded in. Many mathematic models can be found in the literature that evaluate electrochemical fuel cell performance [1–9]; however the model must be developed according to a specific target. In other words, if the main purpose of the study is, for example, shape optimization, the model must be very accurate concerning internal conditions, such as current density, temperature distribution, etc. The models found in the literature [1–3, 5, 9] seem to produce excellent results. However, the aim here is to create a mathematical model for fuel cell stacks, which can be integrated into a simulation modular code for the Balance of the Plant (BoP) optimization. Plant modelling can be performed using commercially available software. Often, since fuel cells are innovative devices, numerical codes that simulate their behaviour must be implemented by using a proprietary code. Previous papers can be found in the literature [4, 10, 11], where fuel cell stacks are inserted in an entire power plant, but very often the fuel cell block is zero-dimensional and it is not affected by the variation in input parameters. The authors, for this purpose, have developed a module to be integrated into commercial software for plants simulation, Aspen Plus. The model adopted is simple, but at the same time, it calculates with good accuracy fuel cell performance under any conditions. Another important element taken into account during the code development is the validation through experimental data. In the present work, the authors have adopted an analytic model for evaluating the electrochemical performance [6] of a fuel cell and results obtained are compared with those found out during the tests on a Molten Carbonate Fuel Cell. Some phenomena not described in the original model are considered and an enhanced model for the cell performance is developed. The model obtained is therefore used to analyse fuel cell behaviour under several operating conditions. These results seem to fit very well with experimental analyses conducted in the test rig of the University of Perugia.