Predicting the performance and emission characteristics of a Mahua oil-hydrogen dual fuel engine using artificial neural networks

ABSTRACT

The current research work aims at developing the Artificial Neural Network (ANN) model to predict the performance and emission behavior of diesel-powered engine. Conventional diesel engine was converted into dual fuel mode using high octane and high cetane fuels. Several experiments were performed with varying high-octane fuel flow rate and constant flow of high cetane fuel. These data were used for training and testing the ANN model. Feed forward back propagation algorithm was used to predict the performance and emission behavior of dual fuel engine. Engine load, high octane fuel flow rate, fuel injection pressure and fuel injection timing were used as input parameters of ANN model whereas BTE, EGT, HC, CO, NO, and smoke were used as output prediction parameters. The accuracy of the predicted values was determined by means of MSE and Regression coefficient. Finally, the study concludes that there is a credible behavior between ANN predicted values and experimental values for performance and emission behavior of dual fuel engine.

KEYWORDS:

• Hydrogen
• Green fuel
• ANN
• Emissions
• Dual fuel engine