Prediction of the WPPO Biodiesel-Fuelled HCCI Engine Using Artificial Neural Networks

This paper presents experimental study efforts to explore the performance and emission characteristics of an existing single-cylinder, four-stroke, water-cooled, direct injection Kirloskar diesel engine was converted into HCCI engine. From the investigation, it was stated that WPPO with diesel results increased the brake thermal efficiency by 42.12% at 413 K inlet air temperature and full load condition. Formerly NO_x were decreased for all blends and later slightly increases but smoke is negligible. However, the CO and UHC emissions are first increased and then decreased for the HCCI operation. The ANN was trained, validated and tested with experimental data sets. The artificial neural network system was created to predict the performance and emission parameters of the engine. A multi-layer discernment network was utilized for non-straight mapping among input and output parameters. Six objectives—BTE, EGT, NO_x, Smoke, CO and UHC were considered. The performance of the ANN model is determined also illustrations the efficiency of the model to predict the performance and emission with a determination coefficient of 0.999.