An insight into the estimation of fatty acid methyl ester based biodiesel properties using a LSSVM model. (1st May 2019)

• Record Type: Journal Article
• Authors: Razavi, Razieh; Bemani, Amin; Baghban, Alireza; Mohammadi, Amir H.; Habibzadeh, Sajjad
• Is Part Of: Fuel. Volume 243(2019)
• Page Start: 133
• Page End: 141
• Publication Date: 2019-05-01
• Subjects: Biodiesel -- Viscosity -- Cloud point -- Iodine value -- Pour point -- LSSVM model; Fuel -- Periodicals; Coal -- Periodicals; Coal; Fuel; Periodicals; 662.6
• DOI: 10.1016/j.fuel.2019.01.077
• Languages: English
• ISSNs: 0016-2361

Abstract

Graphical abstract: Highlights: LSSVM-PSO model has been developed for estimation of fatty acid methyl ester based biodiesel properties. Pour point, cloud point, iodine value and kinematic viscosity are modeled as a function of fatty acids composition. Good predictions are obtained by the suggested approach. Abstract: Biodiesel as an environmental friendly and renewable fuel assigns a great potential to substitution of petroleum diesel. The biodiesel characteristics are highly dependent on the structure of its basis oil. The main aim of the present work was to develop an accurate model based on LSSVM-PSO algorithm to estimate biodiesel properties, i.e., pour point, cloud point, iodine value and kinematic viscosity as a function of fatty acids composition. The temperature, molecular weight, weight percent of saturated acids, poly unsaturated fatty acids and mono unsaturated fatty acids, number of double bonds, and carbon number are involved variables for development of the models. The performance of LSSVM-PSO model is evaluated using different statistical criteria which result in the coefficients of determination of 0.99995, 0.99981, 0.99848 and 0.99930 for pour point, cloud point, iodine value and kinematic viscosity, respectively. In addition, the outcomes of the proposed model are compared with those from an ANFIS model indicating the great potential of the LSSVM-PSO model to estimate the biodiesel properties.