Search Results

This paper explores the feasibility of using in-cylinder pressure-based variables to predict gaseous exhaust emissions levels from a Navistar T444 direct injection diesel engine through the use of neural networks. The networks were trained using in-cylinder pressure derived variables generated at steady state conditions over a wide speed and load test matrix. The networks were then validated on previously “unseen” real-time data obtained from the Federal Test Procedure cycle through the use of a high speed digital signal processor data acquisition system. Once fully trained, the DSP-based system developed in this work allows the real-time prediction of NOX and CO2 emissions from this engine on a cycle-by-cycle basis without requiring emissions measurement.

The effects of fuel composition on emissions levels from compression ignition engines can be profound, and this understanding has led to mandated reductions in both sulfur and aromatic content of automotive diesel fuels. A Navistar T444E (V8, 7.3 liter) engine was installed on an engine dynamometer and subjected to transient emissions measurement using a variety of fuels, namely federal low sulfur pump diesel; California pump diesel; Malaysian Fischer-Tropsch fuel with very low sulfur and aromatic content; various blends of soy-derived biodiesel; a Fischer-Tropsch fuel with very low sulfur and 10% aromatics; and the same Fischer-Tropsch fuel with 10% isobutanol by volume. The biodiesel blends showed their ability to reduce particulate matter, but at the expense of increasing oxides of nitrogen (NOx), following the simple argument that cetane enhancement led to earlier ignition. However, the Fischer-Tropsch fuels showed their ability to reduce all of the regulated emissions.