Skip to Main Content
The operating temperature of PEMFC stack is a very important control parameter. In operating process, electro-chemical reaction and the humidity of proton exchange membrane vary sensibly with it, the variation of operating temperature has a significant influence on the output performance and lifespan of fuel cells. The most existing PEMFC mathematical models are too complicated to apply in on-line control process, so it is necessary to find a simple accurate stable control method. In this paper, a novel genetic algorithm (NGA) is presented, which possesses stronger global searching and local optimizing capacities than SGA. NGA is applied to optimize rapidly the weights of neural network according to the population of excellent individuals, neural networks technique is used to establish a self-adjusted control model for PEMFC system. The results of simulation and experiment are given in the end.