Experimental and Scale-Resolving Simulation Study of Flow Dynamics in Fuel Cell Cooling Systems

The flow dynamics in fuel cell bipolar plate cooling systems are characterized by a sequence of complex flow regimes, including impinging jets, flow redistribution, laminarization within filigree cooling channels and subsequent transition downstream. Accurate prediction of these phenomena is essential for reliable thermal management but remains challenging due to strong turbulence anisotropy and flow transition. In the present study, a combined numerical and experimental investigation of a reference bipolar plate cooling configuration is presented. The numerical analysis is based on the eddy-resolving k–w–z–f (ER–z–f ) turbulence model and is assessed against wall-resolved Large-Eddy Simulations (LES) using the WALE subgrid-scale model. To support the numerical findings, spatially resolved velocity measurements inside selected cooling channels are obtained using micro particle image velocimetry (m-PIV). The results show that the ER–z–f model accurately reproduces the dominant flow features and turbulence dynamics predicted by the LES, including impingement-induced turbulence production, flow redistribution and the progressive laminarization inside the cooling channels. The streamwise evolution of velocity profiles within the channels is in close agreement with the m-PIV measurements, confirming a strong attenuation of turbulent fluctuations and a transition towards laminar flow conditions. Overall, the study demonstrates that the ER–z–f model provides a reliable and computationally efficient alternative to LES for the prediction of complex flow phenomena in fuel cell cooling systems.