Low temperature proton exchange membrane fuel cells (PEMFC) are a promising solution to achieve zero toxic emissions in heavy duty vehicles. Despite progress in performance, efficiency and envisaged lifetime, component degradation under operating
conditions remains a key challenge. This work analyses PEMFC performance at different points of lifespan and develops a methodology to use a model as a virtual sensor to monitor component degradation in a PEMFC. Using experimental data that envelops performance measurements across the entire PEMFC lifespan for different catalyst materials, we calibrated the electrochemical model using the developed methodology of discrete virtual sensing of PEMFC aging and noted the evolution of its intrinsic parameters. The model exhibits good predictability, achieving a high degree of agreement with experimental data with the lowest value of R2 being 0,96709, while simultaneously obtaining uniquely identifiable parameters, confirmed by Fisher information matrix analysis. It also enables a unique view into the amount of information provided from each experimental point. Results show that the proposed virtual sensing methodology successfully detects degradation of three intrinsic parameters, confirmed by comparison with in-situ measurements of the electrochemically active surface area, high frequency resistance, voltage, current and ex-situ measurements of porosity, membrane thickness and other imaging material.
|
