| dc.description.abstract | Estimating emissions of an internal combustion engine depends on the interactions between different vehicle components, such as the three-way catalyst
(TWC) and the exhaust manifold. This thesis focuses on developing a transient integrated low-order model for multi-cylinder spark-ignition engines to predict the formation, transfer, and reduction of significant pollutants NO, CO, CO2, O2, and UHC (unburned hydrocarbons). This was achieved through integrating four models: a torque-speed model, an exhaust manifold model, a combustion model, and a TWC model.
The combustion process is modeled as a continuously stirred-tank reactor (CSTR) and assumes a simplified gasoline formulation using a two-lumped reaction mode. The exhaust manifold is modeled based on the conservation of energy, mass, and momentum equations and presented by a set of three first-order hyperbolic partial differential equations (PDEs). The exhaust gas properties at the manifold are obtained by using the Lax-Wendorff numerical scheme to solve the PDEs. The gas velocity, temperature, and density are estimated throughout the manifold length and operating times. The emission reduction in TWC is predicted by a lumped analysis of reductants and oxidants and accounts for oxygen. It is assumed that the reaction only occurs at the wash coat, and symmetry simplifies the TWC from a three-dimensional to one-dimension model.
The integrated model is coupled with a torque-speed model to convert a predefined vehicle speed profile to the engine's torque and RPM by considering the main powertrain components: flywheel, gearbox, differential drive-axle, and wheel size. The emission prediction with the torque-speed model minimizes extensive emission mapping techniques and optimizes the various vehicle system dependencies
The variations between experimental data and theoretical models are mainly due to their difference in spark timing, fuel composition, heating value, and the exact molecular weight of fuel. The fuel data is imported from literature, where quantifying fuel composition accurately would improve the model's prediction accuracy and improve the overall combined model emissions prediction for SI (spark ignition) engines. The integrated model presented here sites the ground for developing and testing customized driving cycles for future emission regulation purposes. | |
