Modeling and analysis of hydrogen effect on a petrol engine operation using neural networks

Nowadays, The hydrogen engines are widely used in machinery. Adding small and controllable amounts of the hydrogen in the main fuels such as gasoline can be used to complete the combustion, to produce higher efficiency and lower emissions. In this research, a four stroke SI engine operation is modeled using GMDH type of the neural networks and an evolutionary method to make the network structure, so that the percent of the hydrogen are added to the gasoline and then, the brake power and thermal efficiency of the engine are obtained as the functions of the engine speed and the fuel mass flow rate. Based on obtained results, these functions accurately evaluated the operation of the research engine. The modeling results also showed that the brake power has more dependency on the engine speed and the thermal efficiency has more dependency on the mass flow rate.