An AI-based approach to improve the fuel properties and performance of nano graphene added straight Karanja oil prepared with optimal dispersion parameters

Abstract

In this study, a stable nano-graphene oxide added Karanja oil (GKO) was prepared with optimal nano-fluid dispersion parameters. For this, samples were prepared with the selected dosage of graphene oxide (GO) (30–90 ppm), sodium dodecyl sulfate (SDS) with GO:SDS ratio of 1:1 to 1:5 and were subjected to ultrasonication stirring for 30–45 min. The stability of nano-fluid dispersions was decided based on zeta potential measurement. The optimal dispersion parameters were obtained using a well-trained radial basis function (RBF) neural network integrated with the particle swarm optimization (PSO) technique. The predicted optimal nano-dispersion parameters were 58.3-ppm GO, 180.4 ppm SDS and 38 min ultrasonication time. The viscosity of Karanja oil (KO) and GKO was found within the ASTM limit at 105 °C temperature. The fuel properties of GKO, preheated Karanja oil (PKO) and Karanja oil (KO) were measured by following the latest ASTM standards and found to be comparable with diesel fuel. Furthermore, experiments were conducted to investigate the performance of the prepared fuel blends in a 10-kW diesel engine. Preheated GKO showed comparable performance with diesel and showed higher BTE (upto11.81%), lesser BSFC (up to 17.72%) and EGT (up to 4.11%) as compared to KO and PKO, respectively.

Keywords:

• RBF-PSO
• graphene oxide
• nano-fluid dispersion
• zeta potential
• fuel properties

Disclosure statement

No potential conflict of interest was reported by the authors.