An innovative approach to the study of the effect of alternative fuels on the vibration activity of internal combustion engines

BACKGROUND: The use of ammonia (NH3) as a diesel fuel additive is of great interest for addressing environmental and operational challenges, such as reducing harmful emissions and improving engine efficiency. However, the influence of ammonia on vibration parameters of diesel engines is still understudied, thus requiring detailed research. AIM: This study aims to analyze the influence of ammonia (NH3) as a diesel fuel additive on vibration parameters of a diesel engine using an artificial neural network (ANN) designed to predict these parameters. METHODS: The study employs modeling methods based on neural networks. TensorFlow and Keras frameworks were used to develop and train the neural network. Parameters of the engine (engine bore, speed, compression ratio, etc.) and fuel mixtures containing 5%, 10%, and 15% NH3 were used as inputs. The model's accuracy was assessed based on experimental data. RESULTS: The developed ANN showed high vibration prediction accuracy (up to 98.67%). Experimental data showed that increased ammonia additive ratio in the fuel mixture results in increased engine vibration. The highest vibration amplitude was 11.67 m/s² at 15% NH3. The results indicate potential risks of lower engine performance when using ammonia, despite its environmental advantages. CONCLUSION: The study showed that ammonia additives in diesel fuel increase engine vibration, which could affect its reliability and durability. Despite the high accuracy of neural network predictions, the observed increase in vibration requires further investigation. Additional research is required to ensure the safe use of NH3 as a fuel additive by focusing on optimal mixture proportions and the influence of other factors, such as pressure and temperature.