A Virtual Sensor for a Cell Voltage Prediction of a Proton-Exchange Membranes Based on Intelligent Techniques

Abstract

The use of Proton-Exchange Membranes Fuel Cells is presented as a key alternative to face the increasing and concerning problems related to global warming. The international expansion of green policies, has resulted in the need of ensuring their quality and reliability performance. Although fuel cells can get to play a significant role, this technology is still under development, paying special attention to the problems related to gas starvation and degradation. In this context, the present work deals with the virtual sensor implementation of one of the voltage cells present in a stack, whose operation is subjected to several degradation cycles. The proposal predicts indirectly the voltage of one cell from the current state of the rest of the cells by means of an intelligent model.