An experimental study and ANN analysis of utilizing ammonia as a hydrogen carrier by real-time emulsion fuel injection to promote cleaner combustion

Abstract

This work utilized Real-Time Emulsion Fuel Injection (RTEFI) methods to incorporate emulsion fuel, which included citronella oil (20%), ammonia (10%), and water (5%), into diesel fuel without the addition of surfactants. The primary aim of this study was to investigate the influence of hydrogen presence in liquid ammonia on diesel engines. This was accomplished by using a water-emulsified diesel-biodiesel blend and forecasting performance using an artificial neural network (ANN). Citronella oil (20%) was blended with standard fuel (SD), followed by the addition of 10% ammonia and 5% water, labeled as SD20CO, SD20CO10A, and SD20CO10A5W, respectively. The results indicated that the brake thermal efficiency of SD20CO10A5W is comparable to that of diesel. The fuel combination significantly decreased hydrocarbon emissions by 70–75%, smoke by 10–15%, and oxides of nitrogen by 2–5%, while increasing carbon dioxide by 3–8% under all load situations compared to other fuels. The trained ANN models had a coefficient of determination of 97%, with values ranging from 0.9076 to 0.9965. Additionally, the models exhibited low mean absolute percentage error values, ranging from 0.98% to 4.26%. The analytical results provide strong evidence supporting the efficacy of SD20CO10A5W, ultimately indicating that it is a feasible substitute fuel for diesel engines.

Keywords

• Ammonia emulsion
• Citronella oil
• Emission control
• Diesel engine
• Artificial neural network