Prognostic and Energy Management for Multi-Stack Fuel Cell Systems with Stochastic Non-Homogeneous Degradation

This paper presents a load-dependent degradation model for Proton Exchange Membrane Fuel Cells using a nonhomogeneous stochastic gamma process to predict degradation under varying load conditions. The model effectively captures uncertainties, variability, and non-linearities in the degradation process. Building on this foundation, a novel energy management strategy is developed, specifically designed for multi-stack fuel cell systems, combining degradation prognosis with health-conscious energy management for piecewise-static power applications. The methodology is demonstrated on a two-stack PEMFC system, achieving significant improvements in system lifespan and reliability compared to traditional energy management approaches. These results underscore the proposed model's potential to enhance the durability and operational efficiency of multi-stack systems.

Keywords: Proton exchange membrane fuel cells, Stochastic degradation modeling, Prognostics and health management, Energy management, Multi-stack systems.