Experimental investigation and artificial neural networks (ANNs) based prediction of engine vibration of a diesel engine fueled with sunflower biodiesel – NH3 mixtures

Abstract

Ammonia (NH3) has a high utilization potential in the world's fuel sectors. Therefore, it is recognized as an important alternative according to other conventional fuels. Besides, the fuel additive is intensively used to eliminate the deficiencies in engine emission, performance, and etc. This study was experimentally conducted to explore the effects of NH3 as a fuel additive on engine vibration parameter. In the scope of this study, NH3 was blended into the sunflower biodiesel at volumetric ratios of 5%, 10%, and 15%. Besides, all mixtures (B95A5, B90A10, and B85A15) were combusted under no-load conditions in varying engine speed (1200–2400 rpm) by using Mitsubishi Canter 4D31 diesel engine. Experimental vibration data were recorded with the accelerometer sensor placed on the engine block for all NH3 ratios. Artificial neural networks (ANN) model was generated to predict the experimental vibration results. Vibration data in the x-axis, y-axis, R (resultant), time (second), and engine speed (rpm) were considered as the input parameters to predict z-axis vibration data. Therefore, the z-axis was selected as the target parameter. All experimental vibration data was compared with the predicted vibration data and a high accuracy rate has been achieved in this study. As a result of the vibration experiments, the increment was observed in the vibration data when the NH3 additive was gradually increased. In this sense, it has been revealed that the NH3 additive in the sunflower biodiesel negatively affected the engine vibration. An apparent vibration change was observed for all the NH3 additive depending on the varying engine speed. The highest and lowest vibration data were respectively recorded as 12.96 m/s2 and 9.12 m/s2 for 15% and 5% NH3 additive. Besides, the accuracy rates of the generated ANNs models were respectively calculated as 99.206%, 99.675%, and 99.505% for the sunflower biodiesel blended with 5%, 10%, and 15% NH3 additive.