Physically motivated structuring and optimization of neural networks for multi-physics modelling of solid oxide fuel cells

Neural network models for complex dynamical systems typically do not explicitly account for structural engineering insight and mutual interrelations of various subprocesses that are related to the multi-physics nature of such systems. For that reason, they are commonly interpreted as a kind of data-driven, black box modelling option that is in opposition to a physically inspired equation-based system representation for which suitable parameters are subsequently identified in a grey box sense. To bridge the gap between data-driven and equation-based modelling paradigms, this paper proposes a novel approach for a physics-inspired structuring of neural networks. The derivation of this kind of structuring, an optimal choice of network inputs and numbers of neurons in a hidden layer as well as the achievable modelling accuracy are demonstrated for the thermal and electrochemical behaviour of high-temperature fuel cells. Finally, different network structures are compared against experimental data.