Over the years many investigations of the gas temperature at the exhaust valve have been made. Nevertheless the modeling of the gas temperature still remains an unsolved problem. This master thesis approaches the problem by attempting to model the exhaust gas temperature by using the standard sensors equipped in SI engines, together with an in-cylinder pressure sensor which is needed in order to develop certain models.
The concept in the master thesis is based upon a parameterization of the ideal Otto cycle with tuning parameters which all have physical meanings. Input variables required for the parameterization model is obtained from a fix point iteration method. This method was developed in order to improve the estimates of residual gas fraction, residual gas temperature and variables dependent of those, such as temperature at intake valve closing.
The mean value model of the temperature, at the exhaust valve, is based upon the assumption of the ideal gas law, and that the burned gases undergoe a polytropic expansion into the exhaust manifold. Input variables to the entire model are intake manifold pressure, exhaust manifold pressure, intake manifold temperature, engine speed, air mass flow, cylinder pressure, air-to-fuel equivalence ratio, volume, and ignition timing.
A useful aspect with modeling the exhaust gas temperature is the possibility to implement it in turbo modeling. By modeling the exhaust gas temperature the control of the turbo can be enhanced, due to the fact that energy is temperature dependent. Another useful aspect with the project is that the model can be utilized in diagnostics, to avoid hardware redundency in the creation of the desired residuals.
Source: Linköping University
Author: Ainouz, Filip | Vedholm, Jonas