Prediction of CI Engine Performance and Emission Characteristics Powered by WPO and Diesel Fuel Blends by ANN

Abstract

The emissions and operational performance characteristics of a four-stroke compression ignition engine employing a mixture of pure diesel and catalytic pyrolysis oil obtained from municipal plastic waste. In the current study, a framework of an Artificial Neural Network (ANN) is employed to simulate. The maximum amount of oil produced at 350°c - 500°c. The liquid fuel that was produced physical characteristics were like those of diesel, such as density (790 kg/m3) and calorific value (39.6 mj/kg) which helped the fuel blend burn more completely and effectively, improving performance and combustion characteristics. an Artificial Neural Network (ANN) model was built and used to anticipate emission characteristics (smoke, hc, co, nox) and performance metrics including (brake power, brake thermal efficiency and brake specific fuel consumption) for inputs parameters load and volume of selected blends. for the most accurate prediction of emissions and performance parameters the back-propagation algorithm was employed. The regression coefficients (r2) of 0.99801, 0.9983, 0.95753, and 0.97467 for bthe, bsfc and emission, which were extremely close to 1. According to the research work, an unmodified diesel engine might run on a blend of the suggested alternate fuel and diesel blend. Additionally, it indicates that artificial neural networks are helpful in modelling and predicting the emissions or performance of wpo in diesel engines, which may lead to the use of wpo30 fuel in automotive engines. Major Finding: ANN is a powerful tool for modelling the performance parameters of ci engine powered by waste plastic oil-diesel fuel blends. It proved the strong correlation between the predictions made by ANN and actual experimental data in modeling of new fuel in ci engine.