Neural network analysis of proton exchange membrane fuel cell performance under different operation conditions

The neural network can construct relationships between the control factors and responses in the PEMFC. This study proposes a feed forward neural network modelling for parametric analysis of proton exchange membrane fuel cell (PEMFC) performance. The neural network trained with Lederberg-Marquardt algorithm. Numerous parameters affecting the maximum power density of PEMFC are analysed, such as fuel cell operating temperatures, cathode and anode humidification temperatures and operating pressures. Experimental results are presented for identifying the proposed approach, which is useful to get deep knowledge of PEMFC behaviour.