Machine learning algorithms for a diesel engine fuelled with biodiesel blends and hydrogen using LSTM networks

The combustion of fossil fuels is one of the main reasons for the increase in global warming. Exhaust emitted from burning fossil fuels affects both environment and human health. The present study uses non-edible rubber seed oil (RSO) biodiesel with hydrogen in an unmodified diesel engine. The experiments were carried out with 10 %, 20 %, and 30 % biodiesel blends with 10 L/min of constant hydrogen supply. A series of experiments were conducted in the constant engine speed with a varying load of 0–100 % in intervals of 25 %. The hydrogen has been supplied to the combustion chamber via an air intake manifold. Using RSO biodiesel blends with hydrogen increases brake thermal efficiency at lower fraction ratio. Further, the fuel consumption was low when 10 % of biodiesel was used with hydrogen. As the concentration of the blends increased, both brake thermal efficiency and brake specific fuel consumption were massively affected due to the higher viscosity and lower heating values. Due to the effect of heating value, the exhaust gas temperature for the biodiesel blends was higher. The emission of the pollutants such as carbon dioxide, carbon monoxide, hydrocarbons, and oxides of nitrogen was reduced due to the addition of biodiesel and hydrogen blends to the diesel. Considering the findings, the blend with 10 % biodiesel with 10 L/min hydrogen can be a potential alternative for diesel.