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

Research in chemical engineering requires experiments, which are often expensive, time-consuming, and labo-rious. Design of experiments (DoE) aims to extract maximal information from a minimum number of experi-ments. The combination of DoE with machine learning leads to the field of active learning, which results in a more flexible, multi-dimensional selection of experiments. Active learning has not yet been applied in reaction modeling, as most active learning techniques still require an excessive amount of data. In this work, a novel active learning framework called GandALF that combines Gaussian processes and clustering is proposed and validated for yield prediction. The performance of GandALF is compared to other active learning strategies in a virtual case study for hydrocracking. Compared to these active learning methods, the novel framework out-performs the state-of-the-art and achieves a 33%-reduction in experiments. The proposed active learning approach is the first to also perform well for data-scarce applications, which is demonstrated by selecting ex-periments to investigate the ex-situ catalytic pyrolysis of plastic waste. Both a common DoE-technique, and our methodology selected 18 experiments to study the effect of temperature, space time, and catalyst on the olefin yield for the catalytic pyrolysis of LDPE. The experiments selected with active learning were significantly more informative than the regular DoE-technique, proving the applicability of GandALF for reaction modeling and experimental campaigns.

Automated summary

Research in chemical engineering often requires expensive, time-consuming, and laborious experiments. Design of experiments (DoE) aims to maximize information obtained from a minimal number of experiments. Combining DoE with machine learning leads to active learning, allowing for more flexible and multi-dimensional experiment selection. In this work, a novel active learning framework called GandALF is proposed and validated for yield prediction in chemical reactions. GandALF outperforms other active learning strategies, achieving a 33% reduction in experiments in a virtual case study on hydrocracking. It is also the first active learning approach to perform well in data-scarce applications, demonstrated by selecting experiments for catalytic pyrolysis of plastic waste.