Time to market reduction for hydrogen fuel cell stacks using Generative Adversarial Networks

Abstract

In this paper, the objective is to develop a relevant approach to sharply decrease the test time on an experimental test bench, dedicated to conditioning and performance mapping, for fuel cells. In this context, a new concept to reduce the development time of fuel cells by introducing a disruptive and highly efficient data augmentation approach based on artificial intelligence is presented. The proposed innovative concept can support engineering and research tasks during the fuel cell development process to reduce development costs as well as time to market. The results allow for a reduction of the testing time before a product is introduced in the market, from a thousand hours to a few hours.