Nonlinear modeling of MCFC stack based on RBF neural networks identification

Modelling Molten Carbonate Fuel Cells (MCFC) is very difficult and the existing models are too complicated to be used for controlling design, especially for on-line control design. This paper presents the application of neural networks identification method to develop the nonlinear temperature model of MCFC stack. The hidden layer units of the neural networks consist of a set of nonlinear radial basis functions (RBF). The temperature characters of MCFC stack are briefly analyzed. A summary of RBF neural networks for the multi-input and multi-output (MIMO) nonlinear system modelling is introduced. The simulation tests reveal that it is feasible to establish the model of MCFC stack using RBF neural networks identification. The most important thing is that the modelling process avoids complex analytical modelling that uses complicated differential equations to describe the stack. After being tested, the model can be used to predict the temperature responses on-line which makes it possible to design online controller of MCFC stack.