Hydrocracking of complex mixtures: From bulk properties, over fundamental kinetics to detailed product composition

Active learning-based exploration of the catalytic pyrolysis of plastic waste

“…The in-house developed first principle-based hydrocracking model PR1ME was used to develop GandALF and its comparison to the state-ofthe-art [58]. PR1ME consists of two parts, feedstock reconstruction [59] and a microkinetic model [60].…”

“…Next, the novel active learning algorithm is presented, and its performance is compared to other active learning acquisition functions for simulated hydrocracking yields of synthetic naphtha (C 3 -C 9 paraffins). The hydrocracking yields are gathered via PR1ME [45], a first-principle-based hydrocracking model which predicts product yields for complex hydrocarbon mixtures. The comparison of the active learning-framework with other acquisition functions is made with this kinetic model as this allows for quick data generation, as opposed to experimental data.…”