ANN-CFD synergistic approach for improved predictions of pressure evolution in non-venting self-pressurized liquid hydrogen tanks

Abstract Self-pressurization phenomenon due to heat leakage into liquid hydrogen (LH 2 ) tanks, is a primary challenge for long-term storage and safe operation. Accurately prediction of pressure evolution in such tanks remains difficult. Despite the wide-spread use of CFD as a mature-developed non-equilibrium model, intrinsic complex nonlinear phenomena are still hindering its applicability to real practical situations due to assumptions and limitations. Herein, artificial neural network (ANN) as an intelligent modeling approach is first hybridized with CFD model to address the latter model’s shortcomings and improve its predictions by the ANN synergistic effect. This hybrid synergistic approach has been applied to one literature LH 2 test facility where the ullage pressure would be predicted as a response to the hold time under some design conditions. Compared to experimental results, the predictions of the pressure evolution from the hybrid synergistic model has been proven to be desirable in its accuracy with an average error of 0.6 % better than the CFD model solely which presented a significant deviation. This work reveals the strong capability of the ANN-CFD hybrid synergistic model in improving the CFD predictions significantly with robust, accurate and consistent results.