Kinetic model of K–Ni/α-Al2O3 catalyst for oxidative reforming of methane determined by genetic algorithm

[Display omitted] ► K–Ni/α-Al2O3 catalyst was used for methane oxidative reforming at 650°C and 1MPa. ► The reaction network was determined by means of genetic algorithm. ► Direct route was predominant for syngas formation with K–Ni/α-Al2O3 catalyst. ► SVM made a regression model between preparation conditions and rate constants. ► α-Al2O3 diluent accelerated methane combustion to decrease syngas selectivity. Effects of preparation conditions of a K–Ni/α-Al2O3 catalyst on the activity and selectivity of high-pressure reforming of methane was investigated. Catalyst preparation parameters such as calcination temperature of boehmite, the amount of NiO and the amount of K loading were designed by L9 orthogonal array, and the catalysts were prepared by an impregnation method. Each catalyst was used in an activity test where contact time was decreased by increasing the gas feed rate, and conversion was recorded until O2 conversion was below 30%. Both the conversions and syngas selectivity were used for fitting by Genetic algorithm. The algorithm was applied to determine the kinetic parameters of the reaction network of high-pressure reforming of methane. Then a support vector machine was trained using the nine dataset to correlate the catalyst preparation parameters and the kinetic parameters. After the training, we conducted grid searches to build response surfaces of the kinetic parameters. Thus, the all kinetic rate constants could be predicted as functions of the catalyst preparation conditions. The analysis of the kinetic model suggested that successive oxidation of the syngas was the most influential factor for low syngas selectivity. Whereas the amount of NiO loading influences on hydrogen oxidation, CO oxidation was not accelerated by NiO. High syngas selectivity was attained by using a less amount of diluent in the catalyst bed.