Performance and emissions prediction of natural gas engine using two-zone combustion model

Natural gas (NG) is considered as one of the cleanest fuels to replace the conventional fuels such as diesel, petrol because of its rather lower emissions, which has been proven by engine tests at labratories and manufactories in the world. Generally the NG engines share the similar structure profile with diesel or petrol engines, but the combustion characteristics and pollutant emissions are different due to its gaseous properties and higher hydrogen-carbon ratio. In this paper a two-zone simulation model is built to characterize the thermodynamic state of the burnt zone and unburnt zone separately. Each of them is consisted of properties library, mass balance and composition, gas pressure, heat loss and pollutant formation sub-models, etc. The Vibe function, which is based on the first principle of chain reactions, is used to calculate the heat release rate of NG. In order to obtain the gas properties, the in-cylinder gas is considered as a mixture of three well defined basic mixtures - NG, air and stoichiometric gas, whose properties are the functions of temperature. The purpose of mass balance and composition model is to acquire the mass of each zone and proportions of the three basic mixtures. In the gas pressure model, ‘First Law of Thermodynamics’ is used to calculate the in-cylinder pressure. The heat transfer coefficient can be estimated using the Woschni formula while the temperatures of the cylinder wall, cylinder cover, piston crown are assumed to be constant. In pollutant formation model, the ‘Zeldovich Mechanism’ are applied to calculate the NO and concentration in the cylinder. The simulation model is implemented in MATLAB/SIMULINK environment and provides the prediction of engine performance and pollutant emissions. Furthermore, a comparison is carried out between the two-zone model and a single-zone model, in which the power output and NO concentration are chosen as criterions. It is shown in the results that the two-zone model performs better in the prediction of NO emission with a slight difference in power output.