Sulfur oxidative coupling of methane process development and its modeling via machine learning

Sulfur oxidative coupling of methane (SOCM) has seen a significant improvement in catalyst design and performances, but there is still a lack of development at process level. We propose an optimized SOCM process flow diagram, with integrated waste heat recovery system and an efficient separation technique. The outcomes of the simulated process were used to design a data-driven modeling approach, based on machine learning methods, and to evaluate its interpolation accuracy. The simultaneous multi-input/multioutput relationship between the different parameters of the SOCM system were determined, revealing the optimum reaction conditions for the maximum benzene, toluene and xylene yield, at the minimum CH₄ and H₂S recycling rate.