A model for a micro-tri-generation system operated on a CI engine working on various alternate fuels was developed based on the experimental data from multiple fuel blends using an Artificial Neural Network. The experimental data were collected by operating the test setup on different fuel blends for the single-generation and tri-generation. Experimental data input for performance and emissions were taken from a real-sized micro-tri-generation system under four different fuel inputs, i.e., Diesel, Karanja Oil, and Karanja Biodiesel (KB-20 and KB-50). The performance of the Tri-generation working on various alternate fuel sources was comparable with that of diesel as fuel input. The artificial neural network-based model was developed to predict the micro-tri-generation performance and emission using the single generation data. The simulation results showed that the developed ANN 9-16-8 model of the micro-tri-generation system could effectively predict the performance and emission parameters of a micro-tri-generation system working on various alternate fuel blends. The correlation coefficient's values of ANN 9-16-8 model Rtrain, Rval, Rtest, and Rall were 0.9950, 0.9945, 0.9944, and 0.9948, respectively, which showed a higher correlation between the predicted values and the observed values.
All Science Journal Classification (ASJC) codes
- Chemical Engineering(all)
- Fuel Technology
- Energy Engineering and Power Technology
- Organic Chemistry