Multi-domain and Multi-scale model of a fuel cell electric vehicle to predict the effect of the operating conditions and component sizing on fuel cell degradation

Abstract

One of the key challenges in the design of fuel cell vehicles arises from the need for the simultaneous optimisation of their efficiency and lifetime, which is, in particular, challenging when operating under transient operating conditions; further, the minimisation of costs and the use of critical materials should be considered. To enable a more detailed virtual exploration of the design space, this paper presents a high-fidelity multi-domain and multi-scale model of a fuel cell electric vehicle that is capable of modelling the coupled phenomena from the vehicle level to intra-fuel cell level with a mechanistically based fuel cell model. The high-fidelity multi-domain and multi-scale model of a fuel cell electric vehicle is based on consistent coupling of different domains, that is, mechanical, electrical, electrochemical, gas flow, thermal, and control, which are modelled with mechanistic governing equations coupled to empirical degradation models of electrochemical devices and solved with multirate solvers ensuring flux conservations and synchronization of time steps. Establishing a consistent model based causal interactions between different domains and the significantly different scales associated with the phenomena in these domains, which more consistently virtually replicates the phenomena in a real fuel cell electric vehicle, is crucial for evaluating the impact of the parameters from high scales on intra-fuel cell spatio-temporal characteristics impacting fuel cell degradation. The results demonstrate that the application of an experimentally validated multi-domain and multi-scale model of a fuel cell electric vehicle enables the virtual exploration of interactions considering the scale of vehicle characteristics and its driving pattern to the scale of intra-fuel cell spatio-temporal characteristics and its degradation early in the development stages of the fuel cell electric vehicle. Therefore, the proposed approach represents a valuable contribution to the virtualisation of the system-level design of powertrains of fuel cell electric vehicles and supports front loading in the development process.