Search Results

Southwest Research Institute (SwRI) is exploring the feasibility of extending the performance and fuel efficiency of the spark ignition (SI) engine to match that of the emission constrained compression (CI) engine, whilst retaining the cost effective 3-way stoichiometric aftertreatment systems associated with traditional SI light duty engines. The engine concept, which has a relatively high compression ratio and uses heavy EGR, is called “HEDGE”, i.e. High Efficiency Durable Gasoline Engine. Whereas previous SwRI papers have been medium and heavy duty development focused, this paper uses results from simulations, with some test bed correlations, to predict multicylinder torque curves, brake thermal efficiency and NOx emissions as well as knock limit for light and medium duty applications.

SwRI has developed a new technology concept involving the use of high EGR rates coupled with a high-energy ignition system in a gasoline engine to improve fuel economy and emissions. Based on a single-cylinder study [1], this study extends the concept of a high compression ratio gasoline engine with EGR rates > 30% and a high-energy ignition system to a multi-cylinder engine. A 2000 MY Isuzu Duramax 6.6 L 8-cylinder engine was converted to run on gasoline with a diesel pilot ignition system. The engine was run at two compression ratios, 17.5:1 and 12.5:1 and with two different EGR systems - a low-pressure loop and a high pressure loop. A high cetane number (CN) diesel fuel (CN=76) was used as the ignition source and two different octane number (ON) gasolines were investigated - a pump grade 91 ON ((R+M)/2) and a 103 ON ((R+M)/2) racing fuel.

Tremendous amount of useful information can be extracted from the cylinder pressure signal for engine combustion control. However, the physical cylinder pressure sensors are undesirably expensive and their health need to be monitored for fault diagnostic purpose as well. This paper presents the results of the development of a virtual cylinder pressure sensor (VCPS) with individual variable-oriented independent estimators. Two neural network-based independent cylinder pressure related variable estimators were developed and verified at steady state. The results show that these models can predict the variables correctly compared with the extracted variables from the measured physical cylinder pressure sensor signal. Good generalization capabilities of the developed models are observed in the sense that the models work well not only for the training data set but also for the new inputs that they have never been exposed to before.