Pyrolysis is an effective method to valorize plastic waste and obtain value-added fuels. This study adopted the ANN-GA (artificial neural network-genetic algorithm) coupled with a central composition factorial design to optimize the oil production from the pyrolysis of waste polyolefins (WP). The interactive effects of PE mass fraction (20-80 wt%), residence time (20-60 min), and carrier gas flow rate (0-100 mL/min) on the yields of WP pyrolysis products were investigated extensively by ANN. Moreover, the highest WP pyrolysis oil production of 78.87 wt%, optimized by GA, was obtained under 80 wt% PE, 60 min, and 0 mL/min. It was found that the different conditions of PE mass fraction, residence time, and carrier gas flow rate did not change the types of oil's main functional groups (-CH2-, -C=C-, -C=CH2, -CH3, and =C-H). The conditions affected the WP pyrolysis oil fractions significantly. The highest diesel selectivity of 91.42% was obtained under 20 wt% PE, 20 min, and 0 mL/min. Additionally, according to the interactive effects of different conditions on the productions of WP pyrolysis products, the pyrolysis pathways were proposed to understand the pyrolysis mechanism of WP better.
Keywords: Artificial neural network; Diesel-range oil production; Genetic algorithm; Pyrolysis mechanism; Thermal decomposition; Waste polyolefins.
© 2023. The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.