Modelling of Solid Oxide Fuel Cells Applied to the Analysis of Integrated Systems with Gas Turbines

Abstract: Solid oxide fuel cells (SOFCs), working at high temperatures with an incomplete fuel oxidation process, have become an interesting candidate for combination with conventional power generation technology, such as gas turbines, in order to develop power plants that are both functional and efficient. An absolute condition for successful analysis and optimisation of such plants is the existence of reliable simulation tools. Through this research, the goal of creating a method for mathematical simulation of SOFCs suitable for thermodynamic analysis of SOFC/gas turbine hybrid systems has been accomplished. The results is an integrated electrochemical and thermal steady-state model capable of simulating the operating behaviour of a fuel cell, i.e. gas utilisation, power produced, energy efficiency, and current and temperature profiles for different operating conditions. The model was based on the combination of work of leading authors within SOFC modelling. Improvements have been made at single-cell modelling level avoiding many of the assumptions made by earlier researchers who have addressed SOFC modelling in connection with system studies. The improvements are mainly in description of polarisation loss of the cell and also in the heat transfer modelling. The model developed is capable of simulating bipolar, planar cell geometry and also planar tube design. To verify the accuracy of the model predictions for the planar tube design, a comparison was made with single-cell test data from Rolls Royce. Agreement with the electrochemical results is within 2.5% when modelling three-cell modules. For larger twenty-cell modules good agreement was obtained at higher fuel flows. A possible explanation of the disagreement between the results at lower fuel flows and higher current densities could be the uncertainty in the data for the diffusion loss model. For the bipolar planar design a verification test showed good agreement with other models from the literature. Sensitivity studies have been performed elucidate the effect of different assumptions regarding input data, in particular electrochemical reaction and reforming rates, on calculation of the cell performance and on the conclusions drawn. The results showed that the variation in the empirical correlation factors used may lead to significant variations in the calculated temperature and current density distributions and, consequently, different cell performances. To gain a more accurate assessment of the performance of SOFCs, the application of more advanced three-dimensional SOFC models in system studies has also been discussed.