Hybrid Process Models in Electrochemical Syntheses under Deep Uncertainty

Chemical process engineering and machine learning are merging rapidly, hybrid models have shown promising results in analysis design. However, uncertainties first-principles an adverse effect on extrapolations inferences based models. Parameter sensitivities essential tool to understand better the underlying uncertainty propagation system identification challenges. Still, standard parameter sensitivity concepts may fail address comprehensive problems, i.e., deep with aleatoric epistemic contributions. This work shows a highly effective reproducible sampling strategy calculate simulation global for under uncertainty. We demonstrate workflow two electrochemical synthesis studies, including of furfuryl alcohol 4-aminophenol. Compared Monte Carlo reference simulations, CPU-time was significantly reduced. The general findings model studies twofold. First, has significant analysis. Second, predicted add value interpretation themselves but not suitable predicting real process/full model’s sensitivities.