Carbon Monoxide Emissions Model for Data Analytics in Internal Combustion Engine Applications Derived from Post-Flame Chemical Kinetics

WCX World Congress Experience
Authors Abstract
In this work, a new CO emissions formation model is developed based on the dynamics of a representative pool of radicals in the post-flame combustion gases. The ultimate target is the derivation of a kinetics-based CO model, formulated by a single differential equation, that can run very fast in any engine diagnostic/post-processing tool which analyzes in-cylinder processes in the framework of big data acquired at the engine test bench or during engine operation in the field. Specific objectives in the development of the current model are (i) inclusion of the effect of engine operating conditions on the CO emissions formation mechanism, (ii) ease of implementation in any diagnostic code/platform, (iii) fast running times toward real-time capability, and (iv) robustness.
The presently developed CO model consists of a single Ordinary Differential Equation (ODE) that can be solved analytically, without the need of a stiff chemical kinetics solver. A characteristic parameter, which can be considered as a quenched CO index, is also derived to quantify the limits of maintaining the partial equilibrium of CO with the radical pool on the one hand and the quenching of the CO oxidation on the other hand.
The model is used to study the quenching of CO oxidation under lean combustion conditions, as they currently appear in modern Spark Ignition (SI) and Diesel engines. As a first application, characteristic 2D maps of the quenched CO index are produced for a wide range of temperatures and cooling rates of the combustion gases under lean methane/air combustion conditions at various lambda values and pressures, revealing the relevant window of CO oxidation quenching. The CO model is then implemented into a quasi-dimensional, multi-zone combustion diagnostic tool for conventional homogeneous charge SI engines, developed by the authors. The combustion diagnostic tool is applied to a lean burn gas engine over a wide range of engine speeds at full load and contact lambda value. Qualitative results of the CO quenching process in the temperature-stratified burned gas are shown and explained. Finally, validation of the present CO model is performed comparing calculated CO emissions of the model against measured engine-out ones.
Meta TagsDetails
Bikas, G., and Michos, K., "Carbon Monoxide Emissions Model for Data Analytics in Internal Combustion Engine Applications Derived from Post-Flame Chemical Kinetics," SAE Int. J. Engines. 11(6):947-964, 2018,
Additional Details
Apr 3, 2018
Product Code
Content Type
Journal Article