ANN-ANFIS model for optimising methylic composite biodiesel from neem and castor oil and predicting emissions of the biodiesel blend

Researchers and stakeholders have shown interest in heterogeneous composite biodiesel (HCB) due to its enhanced fuel properties and environmental friendliness (EF). The lack of high viscosity datasets for parent hybrid oils has hindered their commercialisation. Reliable models are lacking to optimise the transesterification parameters for developing HCB, and the scarcity of predictive models has affected climate researchers and environmental experts. In this study, basic fuel properties were analysed, and models were developed models for the yield of HCB and kinematic viscosity (KV) for composite biodiesel/neem castor seed oil methyl ester (NCSOME) using Artificial Neural Network (ANN) and Adaptive Neuro Fuzzy Inference System (ANFIS). Statistical indices such as computed coefficient of determination (R2), root-mean-square-error (RMSE), standard error of prediction (SEP), mean average error (MAE), and average absolute deviation (AAD) were used to evaluate the effectiveness of the techniques. Emission models for NCSOME-diesel blends were also established. The study investigated the impact of optimised fuel types/NCSOME-diesel (10-30 vol%), ZnO nanoparticle dosage (400-800 ppm), engine speed (1100-1700 rpm), and engine load (10-30%) on emission characteristics and environmental friendliness indices (EFI) such as carbon monoxide (CO), Oxides of Nitrogen (NOx), and Unburnt Hydrocarbon (UHC) using Response Surface Methodology (RSM). The ANFIS model demonstrated superior performance in terms of R2, RMSE, SEP, MAE, and AAD compared to the ANN model in predicting yield and KV of HCB. The optimal emission levels for CO (49.26 ppm), NOx (0.5171 ppm), and UHC (2.783) were achieved with a fuel type of 23.4%, nanoparticle dosage of 685.432 ppm, engine speed of 1329.2 rpm, and engine load of 10% to ensure cleaner EFI. The hybrid ANFIS and ANN models can effectively predict and model fuel-related characteristics and improve the HCB process, while the RSM model can be a valuable tool for climate and environmental stakeholders in accurate forecasting and promoting a cleaner environment. The valuable datasets can also provide reliable information for strategic planning in the biodiesel and automotive industries.