Development of jet surrogate fuel formulated by genetic algorithm

Abstract

Aviation fuels are complex mixtures containing hundreds of species, making experimental and numerical studies challenging. To simplify these studies, surrogate fuels - simplified mixtures that replicate the essential properties of the actual fuel - are used. In this study, surrogate fuels for POSF-10325, JP-8, and RP-3 were formulated using a genetic algorithm, an optimization technique. A chemical species palette with 16 candidate species was utilized, selecting one representative species from each class (n-alkanes, iso-alkanes, cyclo-alkanes, and aromatics) to optimize both the composition and mole fractions. This approach enables a more flexible and precise surrogate fuel formulation by simultaneously optimizing both chemical composition and mole fractions. By applying a multi-objective function, the optimized surrogate fuel for POSF-10325 exhibited minimal deviations in key temperature-independent properties, with errors of 1 % for molecular weight (MW), -0.4 % for derived cetane number (DCN), 0.5 % for lower heating value (LHV), 0.8 % for threshold sooting index (TSI), and 1 % for the hydrogen-to-carbon (H/C) ratio. Additionally, the distillation curve, a temperature-dependent property, was incorporated into the multi-objective optimization process. The optimized surrogate fuels were validated through comparisons with ignition delay time (IDT) measurements and reactor-based experiments, including plug flow reactor (PFR) and jet-stirred reactor (JSR) tests. The results highlighted the effectiveness of the proposed optimization framework in accurately formulating surrogate fuels while ensuring compatibility with experimental combustion characteristics.