Research Information System
Niğde Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi, vol.12, no.4, pp.1593-1605, 2023 (Peer-Reviewed Journal)
In the literature, multiple linear regression, machine learning methods, and group contribution methods have been employed to estimate the cetane numbers of pure biodiesels based on their properties (composition of fatty acid esters, number of carbon atoms, number of double bonds, chain length, saponification number, iodine value, etc.). However, there has been relatively limited research on the relationship between cetane number and other fuel properties. Therefore, this study purposes to utilize the multiple non-linear regression method to estimate the cetane numbers of pure biodiesels as functions of the density, kinematic viscosity, flash point, and heating value. To establish correlations, experimental data on the fuel properties of 100 different biodiesels (methyl and ethyl esters) were gathered from the literature. The predictive performances of the proposed multiple non-linear correlations were compared with the commonly recommended multiple linear correlation found in the literature. According to the results, reliable non-linear correlations, having relative errors of less than 5% and high coefficient of determination values (r2) were obtained.