The Neural Network analysis for the single cell of Molten Carbonate Fuel cell (MCFC)

In the present work try to trained the performance and evolution for the single cell of the MCFC by using the Neural Network tool in the MAT-Lab software. The data used for the Neural Network training are, simulated results, these are obtained for the single cell of the MCFC [1]. The analysis carried out for n input vectors (known input variables i.e. temperature and load current) and power as output vector. Figure 2 shown simulated powers at the different values of input variables, as load current & temperature. Figures 3 shown the trained results are obtained using model in the form of approximate feed forward neural network for the 4 layers & 2: 3: 2 neurons. Power as the output vector of the MCFC is well compare to the simulated results shown in figure 5.