This study aims to solve the inverse problem of chemical kinetics with the genetic algorithm (GA) as an optimization tool to develop robust gasification kinetic models that target a variety of feedstocks, such as southern pine biomass, lignite coal, municipal plastic wastes, and their blends. A dataset (syngas compositions, tar compositions, and char yield) obtained by conducting oxy-steam fluidized-bed gasification experiments of various feedstocks, including southern pine biomass, lignite coal, municipal plastic waste, and their twelve different blends have been utilized to develop and validate gasification kinetic model. The results generated by the developed GA-based gasification model surpassed the performance of the conventional model (developed using a trial and error approach), with the least average absolute errors of 0.9 % for H2, 1 % for CO, 0.65 % for CO2, and 3 % for CH4 in the syngas prediction. The study contributed by developing individual gasification kinetics models for various feedstocks, including biomass, coal, plastics, and their blends. Furthermore, the model was found to be suitable for a range of gasification operating conditions, such as steam-to-carbon ratios of 1–3, equivalence ratios of 0–0.2, and gasifier temperatures of 750–950 °C.
Read full abstract