State of health (SOH) estimation is essential to ensure safety and reliability in
the operation of Proton Exchange Membrane Fuel Cells (PEMFCs). The aging of fuel
cells results from the deterioration of multiple internal components, and the
aging degree of some key components even directly determines the end of cell
life. Due to the complexity of the internal reactions in fuel cell, many
internal parameters cannot be measured or recorded during aging tests. In
addition, external characteristics do not reflect the internal changes in the
cell. Therefore, establishing a multi-scale metric based on fuel cell components
is very important for fuel cell life research. During the aging process of a
fuel cell, the contributions of different components to the overall aging vary
significantly. Additionally, the allocation of indicator parameters presents a
challenge in multi-scale modeling. To address these issues, this paper proposes
a method to construct multi-scale indicators for fuel cells. Firstly, based on
the operational mechanisms of fuel cells, a 3-D Computational Fluid Dynamics
(CFD) model of the fuel cell is developed using COMSOL Multiphysics 6.2 to
simulate the working environment of the fuel cell. In addition, based on
existing research, the aging mechanisms of various fuel cell components are
analyzed, and aging models are established. The aging of selected components is
then simulated in MATLAB R2023a based on the component aging mechanism model.
Moreover, a co-simulation platform based on COMSOL and MATLAB is established to
facilitate parameter interaction and iteration between the two models, thereby
obtaining the aging data of the cell. Finally, the data is analyzed to select
parameters and allocate coefficients for the multi-scale aging indicators. The
multi-scale aging indicator can provide an effective approach to characterizing
the aging state of fuel cells.