PEMFC IDENTIFICATION USING DEEP LEARNING DEVELOPED BY IMPROVED DEER HUNTING OPTIMIZATION ALGORITHM

Fuel cell is an electrochemical system to convert chemical energy directly and without combustion into electricity. Due to several applications and high energy performance of the fuel cell, it has been recognized as one of the most popular energy-conversion systems. These reasons make optimization of the fuel cell systems as one of the most sought-after energy demand and research issues. This study presents a new method for identifying the proton exchange membrane fuel cell (PEMFC) optimal parameters. To do so, an optimization procedure based on a combination of deep-belief network and a new improved version of the deer hunting optimization algorithm has been proposed. The method is then utilized for developing the performance of the parameters in the PEMFC stack. Different operational conditions have been considered for validating the performance of the proposed algorithm by comparison with some different algorithms. Simulation results showed the superiority of the proposed method to achieve high accuracy for forecasting the PEMFC model parameters.

Parameter identification, proton exchange membrane fuel cell, deeplearning, deer hunting optimization algorithm, improved