Date of Award
Master of Science (MS)
Wagner , John
Vahidi , Ardalan
The goal of this research is to develop a thermodynamic simulation of spark-ignition engine combustion that uses a predictive burn-rate model. Previously done thermodynamic engine simulations in MATLAB are based on a specified burn rate model. Also, the effect of turbulence parameters on the rate of mass burn up is not considered. A predictive burn rate model is necessary to study the effect of different fuels on spark ignition engine combustion. The effect of laminar flame speed and turbulent intensity on combustion is difficult to assess experimental due to the difficulty in the measurement of these two variables. Thus, the aim of this simulation is to thermodynamically model the spark ignition engine cycle taking into account the turbulence parameters and their effect on the combustion process. The simulation can be used for predictive studies of combustion of different varieties of fuels by calibration of laminar flame speed, assuming the turbulent intensity to be a fuel-independent parameter.
The key parameters of interest are the turbulence intensity and laminar flame speed. Both these parameters play a key role in controlling mass burn-up rate. The simulation intends to calculate mass fraction burned profile and turbulence intensity from a predictive combustion model based on the concept of turbulent flame entrainment and diffusive burn-up. An empirical relation for laminar flame speed is used for known fuel (gasoline) which can be later calibrated for variety of fuels. The simulation calculates the mass fraction burned profile based on calculated cylinder pressure, temperature and species concentration.
The cylinder pressure and temperature predicted by the simulation match well with the experimental data for the engine used for calibration. The MFB10, MFB50 and MFB90 crank angle positions simulated by the predictive combustion model were compared with those calculated from a commercially available engine thermodynamics program (AVL Concerto) which is based on experimental data and were found within 5-10% of each other. The turbulence parameters, laminar flame speed and the length scales during combustion also showed results in accordance with those shown by previous experiments / simulation.
The quasi-dimensional simulation is done in MATLAB for computational speed and acceptable accuracy. Using MATLAB also makes inserting additional sub-models easier to make the model more detailed and accurate. This simulation will serve as a student guide to engine modeling and also can be used as a base for more advanced simulations.
Desai, Akash, "Laminar Flame Speed Estimation from Experimental Data Using a Quasi-Dimensional Turbulent Flame Entrainment Combustion Simulation For Spark Ignition Engines" (2011). All Theses. 1281.