Nonlinear modeling of a SOFC stack based on a least squares support vector machine

This paper reports a nonlinear modeling study of a solid oxide fuel cell (SOFC) stack using a least squares support vector machine (LS-SVM). SOFC is a nonlinear, multi-input and multi-output system that is hard to model by traditional methodologies. So far, most of the existing models are based on conversion laws, which are very useful for cell design. However, they are too complicated to be applied to control system design. To facilitate a valid control strategy design, this paper tries to avoid the internal complexities and presents a black-box model of the SOFC based on LS-SVM. The simulation tests reveal that it is feasible to establish the model using LS-SVM. At the same time, the experimental comparisons between the LS-SVM model and radial basis function neural network (RBFNN) model demonstrate that the LS-SVM is superior to the conventional RBFNN in predicting stack voltage with different fuel utilizations. Furthermore, based on this black-box LS-SVM model, valid control strategy studies such as predictive control, robust control can be developed.