Artificial Neural Networks for Prediction of Efficiency and NOx Emission of a Spark Ignition Engine

The objective of this paper is the prediction of efficiency and NOx emission of a Spark Ignition engine based on engine design and operational parameters using artificial neural networks (ANN). This paper deals with quasi-dimensional, two-zone thermodynamic simulation of four-stroke SI engine fueled with biogas. The developed computer model has been used for the prediction of the combustion and emission characteristics of biogas in SI engines. Predicted results indicate that the presence of carbon dioxide can reduce oxides of nitrogen (NOx) emissions, but since lower cylinder pressures result, engine power and thermal efficiency are reduced. This is mainly due to the lower heating value of biogas. Using the results from this program, the effects of operational and design parameters of the engine were investigated. For real time computations in electronic control unit (ECU) an artificial neural network (ANN) model has been suggested as an alternative to the engine simulation model. Engine parameters such as equivalence ratio, compression ratio, engine speed, CO₂ fraction in biogas and spark plug position (from the center of cylinder head) and their corresponding engine performance such as efficiency and NOx emission are generated from the developed engine model. These dataset are used for training and testing the ANN model. The validity of these models has been carried out with experimental dataset obtained under same engine setup and yields satisfactory agreement with the predicted values.