Prediction of Fuel Properties of Biodiesel Using Two-Layer Artificial Neural Network

Solomon Olasunkanmi ODEYALE; Adeleye Oladiran EGUNLETI

Authors

Solomon Olasunkanmi ODEYALE Olabisi Onabanjo University, College of Engineering and Environmental Studies, Ibogun, Ogun State https://orcid.org/0000-0002-4479-3992
Adeleye Oladiran EGUNLETI Olabisi Onabanjo University, College of Engineering and Environmental Studies, Ibogun, Ogun State

Keywords:

Biodiesel, artificial neural network, fuel properties, alternative fuel, fatty acid

Abstract

Biodiesel is an alternative fuel produced from a renewable source (biological). Biodiesel has properties similar to diesel, produced from fossil fuels, and this makes it a good substitute as fuel used in diesel engines. The experimental determination of various properties of biodiesel is costly, time consuming and also a tedious process. In order to reduce these problems, researchers have identified that the fatty acid composition of biodiesel determines its fuel properties. The percentage composition of the fatty acid content of each biodiesel plays a significant role, and it is the sole determinant of the fuel properties. Furthermore, artificial neural networks have been considered to be tools helpful in estimating these properties from the fatty acid composition of the fuel. In this study 4 properties (cetane number, flash point, kinematic viscosity, and density) have been modeled, using artificial neural network (ANN). These fuel properties were predicted using 5 fatty acids as input parameters. A 2-layer neural network was used with logsig and purelin in the hidden layers; the fatty acids considered as input parameters were; palmitic acid, stearic acid, oleic acid, linoleic acid and linolenic acid. Both cetane number and flash point used 6 neurons in the hidden layers, and the network had a determining factor (R²) of 0.93488 and 0.9814 respectively. The network of the kinematic viscosity used 7 neurons in the hidden layer, and had a determining factor (R²) of 0.83238, while the density network used 5 neurons and had a determining factor (R²) of 0.819. The results obtained from this study closely agree with previous studies.

doi:10.14456/WJST.2015.26

Downloads

Download data is not yet available.

Metrics

Metrics Loading ...

References

MJ Ramos, CM Fernández, A Casas, L Rodríguez and Á Pérez. Influence of fatty acid composition of raw materials on biodiesel properties. Bioresource Technol. 2009; 100, 261-8.

K Pramanik. Properties and use of jatropha curcas oil and diesel fuel blends in compression ignition engine. Renew. Energ. 2003; 28, 239-48.

WMJ Achten, L Verchot, YJ Franken, E Mathijs, VP Singh, R Aerts and B Muys. Jatropha biodiesel production and use- Review. Biomass Bioenerg. 2008; 32, 1063-84.

GEl Diwani, NK Attia and SI Hawash. Development and evaluation of biodiesel fuel and by-products from jatropha oil. Int. J. Environ. Sci. Tech. 2009; 6, 219-24.

A Demirbas. Mathematical relationships derived from biodiesel fuels. Energ. Sourc. 2008; 30, 56-69.

LC Meher, DV Sagar and SN Naik. Technical aspect of biodiesel production by transesterification-A review. Renew. Sustain. Energ. Rev. 2006; 109, 248-68.

AZ Abdullah, N Razali, H Mootabadi and B Salamatinia. Critical technical areas for future improvement in biodiesel technologies. Environ. Res. Lett. 2007; 2, Article ID 034001.

K Cheenkachorn. Predicting properties of biodiesel using statistical models and artificial neural networks. In: Proceedings of the Joint International Conference on Sustainable Energy and Environment, Hua Hin, Thailand, 2004.

A Mariod, S Klupsch, IH Hussein and B Ondruschka. Synthesis of alkyl esters from three unconventional Sudanese oils for their use as biodiesel. Energ. Fuel. 2006; 20, 2249-52.

G Knothe, JV Gerpen and J Krahl. The Biodiesel Handbook. AOCS Press, Urbana, 2005.

E Alptekin, M Canakci. Determination of the density and the viscosities of biodiesel-diesel fuel blends. Renew. Energ. 2008; 33, 2623-30.

G Knothe. Dependence of biodiesel fuel properties on the structure of fatty acid ethyl esters. Fuel Process. Tech. 2005; 86, 1059-70.

M Canakci and H Sanli. Biodiesel production from various feedstocks and their effects on the fuel properties. J. Ind. Microbiol. Biotechnol. 2008; 35, 431-41.

S Fernando, P Karra, R Hernandez and SK Jha. Effect of incompletely converted soybean oil on biodiesel quality. Energy 2007; 32, 844-51.

AS Ramadhas, S Jayaraj, C Muraleedharan and K Padmakumari. Artificial neural networks used for the prediction of the cetane number of biodiesel. Renew. Energ. 2006; 31, 2524-33.

S Baroutian, MK Aroua, AA Raman and Nik Sulaimon. Estimation of vegetable oil based methyl esters biodiesel density using artificial neural network. J. Appl. Sci. 2008; 8, 3005-11.

J Kumar and A Bansal. Application of artificial neural network to predict properties of diesel biodiesel blends. Kathmandu Univ. J. Sci. Eng. Tech. 2010; 6, 98-103.

X Wei, A Kusiak and H Rahil. Prediction of influent flow rate: a data-mining approach. J. Energ. Eng. 2013; 139, 118-23.

A Verma, X Wei and A Kusiak. Predicting the total suspended solids in wastewater: a data-mining approach. Eng. Appl. Artif. Intell. 2013; 26, 1366-72.

A Kusiak, X Wei, A Verma and E Roz. Modeling and prediction of rainfall using radar reflectivity data: A data-mining approach. IEEE Trans. Geosci. Rem. Sens. 2013; 51, 2337-42.